earlyindicators.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lead-lag relationships between alt-BTC and stock-crypto pairings\n",
    "\n",
    "\n",
    "# Abstract \n",
    "\n",
    "\n",
    "We thought Bitcoin price movements would lead to similar price movements in altcoins. Investors are most confident in Bitcoin as the least risky crypto. For example, imagine a crypto bear market. As Bitcoin recovers, investors gain confidence to put money in riskier coins so as a result, altcoins may lag behind Bitcoin by a certain amount of time. We wanted to perform cross correlation to find that shift. \n",
    "\n",
    "However, using cross correlation, we found the max correlation with a timeshift of zero using 1 minute and 1 day data spanning multiple years. However, we found that on certain days, there are clear lead-lag relationships between BTC and other altcoins or stocks. To this, we developed a trading strategy that capitalizes on these correlations using Linear Regression slope differences as our entry and exit signals.\n",
    "\n",
    "\n",
    "# Introduction:\n",
    "\n",
    "We started with the intent of exploring the relationship between cryptocurrencies. In particular, we expected bitcoin to be a leading indicator of altcoins as it is the largest and most popular crypto. However, we quickly realized that to be minimal, if any. Hence, we shifted to a correlation between crypto assets and stock primarily. Our goal was to be able to trade off of a shift in some indicator that we identify early in the day between a set of possible indicators and assets that can follow those. \n",
    "\n",
    "\n",
    "# EDA \n",
    "\n",
    "We used the yfinance library, so there wasn’t a need to do much data cleaning for EDA. The downside was that the minute data is limited to only a short period of time for download. \n",
    "\n",
    "We looked at BTC and QQQ 5 day price histories to see if they were correlated. They indeed looked correlated. We then normalized the closing prices and performed cross correlation and found BTC to be ahead of QQQ by a number of minutes. We then applied that shift to the price chart.\n",
    "\n",
    "\n",
    "# Methods\n",
    "\n",
    "We realized an issue of cross-correlation is that it is influenced a lot by magnitudes. We care a lot more about the slopes.\n",
    "\n",
    "Our idea: SLOPE LOSS. Use the difference in slopes of the regression line in the loss function. Hyperparameters: stride of slope intervals, length of slope interval.\n",
    "\n",
    "Our method was centered around performing cross-correlation to see the shift in time that two assets have. So, once we identify a shift in the earlier portion of a day’s data and see the quality of that shift, we can trade off of that for the rest of the day. Some things we could do in the future are vary the level at which we want to trade and such, but for the purposes of this project, we focused on implementing the strategy. In order to evaluate the success of a shift, we used something we came up with to find the loss based on the slopes. That requires some hyperparameters for interval length and stride length, which refer to how long we take a local slope over and stride length refers to how spaced apart start points are. We took numbers that seemed reasonable for this as 25min interval length and 15min stride length [stride length < interval length allows for overlap]. There is a sample graph for this in the slides.\n",
    "\n",
    "Our trading strategy is as such: wait for the first half of the day until we have enough bars and perform cross correlation to find a the time shift. If such a shift exists, use the shift to trade. If the difference in slopes is greater than a predefined threshold (hyperparameter), and the asset that is leading has the greater slope, then buy. Do the same but in the downwards direction for selling.  \n",
    "\n",
    "\n",
    "## Results\n",
    "\n",
    "Overall, the trades were mostly positive, winning 52% of the trades made from our aforementioned strategy. Our strategy outperformed BTC heavily during the past 30 days, returning +66% (assuming each trade used 100% of account) while BTC returned -16%. \n",
    "\n",
    "Profit factor was 1.1 which is good considering the state of the market overall and that our strategy only went long. We considered a number of indicators that could work and that led to a working strategy. There are additional optimizations we can do, like online updates to cross correlation as data is coming in. Then, we still would want to try out additional hyperparameter tuning and try to limit losses by placing limits on the slope loss for indicators that we consider to be valid.\n",
    "\n",
    "Moreover, given the downtrend in the current market, it would be better if our strategy could play both long and short. This would just mean adding an additional entry parameter which the inverse of the long entry. Looking at the results by day, we can see a clear positive correlation between BTC's % change that day and our strategy's profitability. Thus, we hypothesize that adding a short entry signal would eliminate this long bias and make our strategy much more profitable in the current market environment. \n",
    "\n",
    "Adding on to this, again looking at the results by day, we can see that there are clearly days when the strategy works well (winrate >60%, avg win >> avg loss on those days) and when the strategy performs poorly (40% or less winrate, avg loss >> avg win). In the future, it would be good to add some way to filter the signals so that trades that are less likely to be profitable can be ignored. \n",
    "\n",
    "However, when looking at trades by token, we can see that QQQ-crypto pairings did not generate any edge. The best performing pairings were BTC-alt or alt-alt. Also, we did not account for slippage or fees in our backtesting. We don't think slippage would be affect profitability very significantly; however, there are no fee-free crypto exchanges currently, so fees would definitely hurt a high frequency strategy like this one.\n",
    "\n",
    "\n",
    "## Conclusion\n",
    "\n",
    "The results of this study were promising. We were able to generate a profit trading only long during a period of time when the markets were extremely bearish. Thus we can reasonably conclude that there does seem to be a significant edge in cross correlation trading. Our pilot study however was very brief and only scratched the surface. There is a lot more research that can and should be done on this topic, including examining different stocks and pairings, adding more and tuning existing hyperparameters, adding more features and indicators that could have predictive power to act as signals or filters to trigger/ignore trades, and more. \n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import yfinance as yf\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from scipy import signal\n",
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n"
     ]
    }
   ],
   "source": [
    "# Get the data\n",
    "qqq = yf.download(tickers=\"QQQ\", period=\"2d\", interval=\"1m\")\n",
    "btc = yf.download(tickers=\"BTC-USD\", period=\"2d\", interval=\"1m\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                             Open_qqq    High_qqq     Low_qqq   Close_qqq  \\\n",
      "Datetime                                                                    \n",
      "2022-04-29 15:55:00-04:00  313.109985  313.420013  312.899994  313.100006   \n",
      "2022-04-29 15:56:00-04:00  313.089996  313.239990  312.809998  312.940002   \n",
      "2022-04-29 15:57:00-04:00  312.940002  312.940002  312.649994  312.747009   \n",
      "2022-04-29 15:59:00-04:00  312.940002  313.559998  312.600006  313.239990   \n",
      "2022-04-29 16:00:00-04:00  313.250000  313.250000  313.250000  313.250000   \n",
      "\n",
      "                           Adj Close_qqq  Volume_qqq      Open_btc  \\\n",
      "Datetime                                                             \n",
      "2022-04-29 15:55:00-04:00     313.100006     1012163  38235.535156   \n",
      "2022-04-29 15:56:00-04:00     312.940002      665858  38236.656250   \n",
      "2022-04-29 15:57:00-04:00     312.747009      854921  38260.964844   \n",
      "2022-04-29 15:59:00-04:00     313.239990     2119364  38301.105469   \n",
      "2022-04-29 16:00:00-04:00     313.250000           0  38310.246094   \n",
      "\n",
      "                               High_btc       Low_btc     Close_btc  \\\n",
      "Datetime                                                              \n",
      "2022-04-29 15:55:00-04:00  38235.535156  38235.535156  38235.535156   \n",
      "2022-04-29 15:56:00-04:00  38236.656250  38236.656250  38236.656250   \n",
      "2022-04-29 15:57:00-04:00  38260.964844  38260.964844  38260.964844   \n",
      "2022-04-29 15:59:00-04:00  38301.105469  38301.105469  38301.105469   \n",
      "2022-04-29 16:00:00-04:00  38310.246094  38310.246094  38310.246094   \n",
      "\n",
      "                           Adj Close_btc  Volume_btc  \n",
      "Datetime                                              \n",
      "2022-04-29 15:55:00-04:00   38235.535156    58343424  \n",
      "2022-04-29 15:56:00-04:00   38236.656250    13608960  \n",
      "2022-04-29 15:57:00-04:00   38260.964844    23250944  \n",
      "2022-04-29 15:59:00-04:00   38301.105469    72230912  \n",
      "2022-04-29 16:00:00-04:00   38310.246094     5863424  \n"
     ]
    }
   ],
   "source": [
    "df = qqq.merge(btc,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime\", suffixes=[\"_qqq\",\"_btc\"])\n",
    "print(df.tail())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "y1 = df['Close_qqq']\n",
    "y2 = df['Close_btc']\n",
    "y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "y2 = (y2 - np.mean(y2)) / np.std(y2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1c2f09f0>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAD4CAYAAADxeG0DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAABQS0lEQVR4nO2dd3gc1dWH37sr7ar3LlmW5N47rhhsY7AJvYTeiYkJkBBCQi8fJSQhJBAcQg0l9G6KMcUG27hhuRu5y5Zl9V5Xq9XO98fdpt5Gkle67/PomdmZ2Zk79u5vz5x7itA0DYVCoVB4L4a+HoBCoVAouocScoVCofBylJArFAqFl6OEXKFQKLwcJeQKhULh5fj0xUWjoqK0lJSUvri0QqFQeC3p6elFmqZFN93eJ0KekpLCli1b+uLSCoVC4bUIIY62tF25VhQKhcLLUUKuUCgUXo4ScoVCofBy+sRHrlAoFHpQX19PdnY2Foulr4eiK35+fiQlJeHr69uh47st5EKIQcDrQCygAS9omvZ0d8+rUCgU7ZGdnU1wcDApKSkIIfp6OLqgaRrFxcVkZ2eTmpraoffo4VqxAXdomjYamAH8RggxWofzKhQKRZtYLBYiIyP7jYgDCCGIjIzs1FNGt4Vc07RcTdO2OtYrgQwgsbvnVSgUio7Qn0TcSWfvSdfJTiFECjAJ2NTCviVCiC1CiC2FhYV6XlahaJGCCgvLd+SgSjUr+ju6TXYKIYKAD4HfaZpW0XS/pmkvAC8ATJ06VX2zFD2K3a5xzrM/kldhIdBkZMGo2L4ekkLRY+hikQshfJEi/qamaR/pcU6FojsUVNaRVyF9jG9sbDEZTqHoN+gRtSKAl4EMTdOe6v6QFIruU1ApRTwqyMz2Y2VomtYvfamKvuexxx7jtddeIyYmhkGDBjFlyhTmzZvH9ddfD8Dpp5/OihUr2L17N7W1tVx33XXs2LGDkSNHkpOTw7Jly5g6dWq3xqCHa2U2cBWwSwix3bHtHk3TvtTh3ApFlyioqANg4ehY3t6cxZHiGlKjAvt4VIqe5OHP9vBzTjOvbrcYnRDCg2ePaXV/eno677zzDtu3b8dmszF58mSmTJnCddddx7PPPsvcuXO58847Xcc/99xzBAQEkJGRwc6dO5k8ebIu49QjamWdpmlC07TxmqZNdPwpEVfohqW+gfLa+k69p6BSCvm8EbJQ3L48fb/gCgXA2rVrOf/88wkICCAkJIRzzjkHgLKyMubOnQvAVVdd5Tp+zZo1XHnllQCMHz+e8ePH6zIOldmpOOH51etbWHugiMw/n9lh94jTtTI2MdTxuq7Hxqc4MWjLcu7vqForihOetQeKANjTwmNzna2B1zccwVLf4NpWUGEhq6SGiEATsSF+GITb1aJQ6MncuXP55JNPqK2tpbKyks8++wyAsLAw1q1bB8Cbb77Z6Pi33noLgN27d7Nz505dxqEscsUJT2yImfyKOlbvLXBZ2E5eXHOYJ7/ej9Vm58aT09A0jZMe/w6ACUmhGA2CqCCzy0JXKPRk8uTJXHLJJUyYMIGYmBimTZsGwH//+1+uv/56hBCcfvrpruOXLl3Kddddx6hRoxg1ahRTpkzRZRzKIlec8DjzeXYeL2+27/OduQCsOVBEna2BK15y56JddlIyADEhZuVaUfQY9957L/v372fdunUMHz4cgClTprBjxw62b9/OzTff7DrW39+fd955h4yMDD766CP8/f11GYMScsUJjd2uUVJtBWB3EyGvqrOxL78SgDX7C/nbV/tYf6gYgJtPHcLFUwcBEBPsp1wrin6Ncq0oTmgqLPXY7BrxoX7kllsoqbYSEWgCYM/xcjQNzhwXx5e78nhpXabrfb87bThGg5wYjQ1RseSK3uGhhx5qti0lJYXdu3e3ePz333+vy3WVRa7oeeqq4F9TYXfnk36LqqQ1flJqBACZRdWufduPlQHwS4fl7WT5LbMx+bg/2iNigymptpKvrHJFP0UJuaLnObIWig/AB9dBfecmHQ8VVgEwNUUK+XPfH+Kpb/ZzsKCKJ7/ex8i4YGYOiWz0nnFNJkTHJYUBsCO7rGvjVyhOcJSQK3oOSwVs+x+8fal725q/deitmqbx9uYsHlq+h0ER/pw3MQGAbzPyeea7A1zzymbqGzROHRGD2cfIpnsWuN7b1H0yJiEEH4Ngg8N/rlD0N5SQK3qOH/4Cn/4GguNh7h9h8Bw4/H3Lx5YchuW3wrp/QkUuxT88z0MfpZNbbuGfl0wi2K9xy6vjZbUYBCw9ZQgAsSF+mH0MRDr85574+Ro5e0IC7/50jPKaesY88BW3vr1N55tVKPoOJeSKniN3B4QNht/thvn3QuxoKNzrjif0ZNubsPV1+PZBeGokUd//ib/7/oe3fzWDKYPDAXjgrNHcvXik6y33nDmK0AC3wG97YCFr/zSvxaFcMT2Z2voGfjxURLW1gc925Oh7r4oByZEjRxg7dmyz7f/85z+pqanptXEoIVf0DJoGebtgyDwwOoKjokeCtQrKsqC6qPHxuTsgZgxc/r5r01nGjZy0fB588yBoGtfPSeWmU4YQFSSt7mGxwY1OEWDyIcDUciDWhEFhBJiMfL7TLeCq4YSip1BCrugfZG0ESxnEelgr8RPl8unx8LchsOoxsDo+7Hk7IX48DD+dO+Ne4dS6v1Ol+WEsPwo//hMeDoPtMrX5w6WzuHByEtNSwjs8HF+jgVlDIvlyV55r24bDxdz78S4a7ErQFV3HZrNxxRVXMGrUKC666CKeeeYZcnJymDdvHvPmySfEr776ismTJzNhwgQWLFjQzhk7j4ojV/QMK+6E0GQYfa57W+JkmP5r2PQf+XrNXyEsWVrtVfkQP5FaawNf5QZRqfkxse4FDv7fQng8Xh7/yVIwBTJ49Ln8/ZcTGl+vMh82PQfTboTQpBaHdPO8oXybUeB6ffmLMgv0hjmppEUH6Xbrij5ixV3yKVBP4sbB4ifaPGTfvn28/PLLzJ49m+uvvx6r1UpCQgKrV68mKiqKwsJCfvWrX7FmzRpSU1MpKSnRd4woi1zRE9SWQt5umHwVBMW4twsBi/8Cd+yDex2WcfEByFwr11Nm89WeXCrrbDx49mg+vuVUMAVAeKrcHzUCvrob7A00Y9d7sO4f8L+LGm+vr4Utr0B1MZOTW7bga6wtnE+h6CCDBg1i9uzZAFx55ZWuYllONm7cyNy5c0lNlZ/jiIgI3cegLHKF/mRtAjRIntny/uA4uYwaDvtXwvpnwT8CYsbwxkcbSIkM4JqZKRgcmZnc+C3UVUDONvjgejiyDtJOaXzO4kNyWZghhXuq7M7C57fDjrdh0wtw8wb+uGgEr6zLZGRcCOsOSj99Z2udK05Q2rGce4qm4a59kT2sLHKFvlQXS8vYFARJ7bSvCk+VUSxCwNlPs/14BVuzyrhmloeIAwRGQUQaDF8MRjMc+Lr5uUoOQeQwCIqFz38PNSWQ/7MU8YAoKfBHf+TmU4ey5b6FnD7G3Yz5o63HqW+wu1432DU1EaroMFlZWWzYsAGAt956izlz5hAcHExlpawDNGPGDNasWUNmpiwhoVwrihOfz26DYxth+Bng205ltxCZ5LN33J3YR57NC2sOEWz2cRW7aoYpAAbPgn1fuidJnRQfgsQp8Ms3AA0OfCOTj3wDYclqMIfAu1dKcQcWjYlzvfXDrdn8e/Uh1+sh93zJ1a9s7vStKwYmI0aMYNmyZYwaNYrS0lKWLl3KkiVLWLRoEfPmzSM6OpoXXniBCy64gAkTJnDJJZfoPgblWlHoR30t7P1crs+/v/3jT/kT6w2TuXxtJBNy1rPjWBm/O20YQeY2PpbTbpSC/N3D0t8OYKuDiuMQkSrF3D8cPl4i9826TU6oXvclvHEBrPgjXPs5MSF+fH37XE7/xxrAXQrAibOZhULRFikpKezdu7fZ9ltvvZVbb73V9Xrx4sUsXry4x8ahLHKFfuz/Si6v/EiKKjJW+3ATkXQREs96n+kA7DhWRlK4P7fOH9b2NUadBWMvlC4TZ92WqnzH+RJkzPrVy2HunXDBS7DgAbkvbpx8b94uV0LS8CZx6ADVdbaO369CcYKghFyhH1tekSGHaadSVWejvKaeRz7PYP7ff3DVEi+qalyBMCNXtm+LDjZzu0fp2TaZfDVYymHnO/J1pSMCJtgRphg/HubfB+MvBqNHan/0SBnbXlVAU46V1vDxtmzVgELhlSjXiqJzaJoM/zP6kJFbwc85FVw4JUlmah5ZByf/AU0YmPOXbyircUeDbMos4YP0bF5df4QXrprC6Q4f9c+5FZwzIYGnL53Y8dn+1LkQPwHSX4Mp10Kl7BLkioZpjegRjsE8B6c91GjXtqwytmWVcc+ZsgSAZxlcxYlNf6wz39nJdvVp7SlKMptPyPUHPrlZJuhYKjjz6R+Ys3w22ponZdalZqc0+XS2HStrJOIAj3z+M6+uPwLICoa2Bjv3f7Kb3HILk5LDOvdFFAJSToaCDLDbm1vkrRE3Xi7X/QM+/jXUVfLWr6Zz9czBrkP25spIg6gWim8pTjz8/PwoLi7uV1FGmqZRXFyMn59fh9+jLPKeoK4SnpkI4y+BC17o69Hoyw6ZJs8Tg/jYlEasKINVj6AhWNswjpvfKObmee7Em7sXj+T1DUc5Xlbr2vbelmxigv14Y+NRAKaldCFBImo42GqhPEta5AZfGYveFgERcNNaeP5k6WOPn8CsGUuZmRZJRW09n2zPcdUsjwhSQu4NJCUlkZ2dTWFhYV8PRVf8/PxISmo5Q7kllJDrjd0On94i149v7dux9AThqVCaCT7+TLQddm8XBm6rv4UqGlixK4/oYDOvXjeNUY7EG6eQnzsxgU+35/Ds6oMABJt9GBnXfNKxXaIdVRDX/VPWdQlPAUMHHjDjx8OIM2UI41d3wZ6PEZOv5okLL+OT7TkcKpQdiNqMnFGcMPj6+royJgcyyrWiN0fXwc+fyPWodiIwvBFbHUy8Au7L45K6+/mP7SzKE+by8ZBHKEMK8q7j5aRFBTImIRSDQTB7aBQAp46I5ulLJzF7qOzoc8X0ZLY9sBAfYxc+hjGjZIx4+n9lss+iP3f8vZe9Ddd+CWMugGObYO1T+PkaGx1it7fyXoXiBEQJud4UZLjXLRV9N46eQNOgphgCIrE12NmkjeIJ2+Wsmf48jx8ZwYKRMfj5yo9UWnSg621nOCY2xztasPk4LOeTUiO6JuIAfiHwh/0w7mJY8CAMW9i596fMhov/K+PdSw7JTFAHadGBNPQjn6ui/6OeH/XEUg6bnge/MEiZI7ve9CesVdBQB4FRrqbIAD8eLKKoqo5zJibw3V4Z2jfWo29malQgq+44haTwAADuWjwSu6axYFQs3cIcBBe+1L1zDDpJLjN/AKRffFB4AGUdqL9yvKyWyEBTM2teoehtlEWuJ5tfkNad3SazC2v0r6nQG7QaAeBsBhEQ1SgTcvU+Kd4TksKYNyIagAsmNZ6oSYsOcoX0jYoP4Y0bpp8YfujkmRAxBNb8nY9unsWbN07HaBA0tONbsdrszH5iFbe/u713xqlQtIEScj1pcFhxZzwOAZFQW9JyW7MTmIeW72Hmn1e1KOb5uVkAbC0ycMVLspZ3Ypg/+RV1BPv5MDgygGcvn8xP956Gv8lLrFSjL0y7AfJ3MTmsltlDozAaBLaGtv/fDhTIMMUVu/PaPE6h6A2UkOtJQz0YfGDKNTLUrcEq3REnCgV74ZlJsOJPrR7y6voj5FVYWN+04/yhVfi8dwX1+PBtsQwvHBoTxJK5aQDMHxmDEIJAsw/RweYeu4UewVlu9+h6AHwMAns7P8B7cvrZ/IfCqzkBnm37EbY68HEE8TtjmmtKwNyF8LqeIG+n9Ntv+o+74FQTRseHEJS3ia83+TF7qKMlVW0p2rtXUaiFc4X1XqqOmvAxWHjrxulEBZkZmxjKxEFhvXcfehM3Xv5/fXU3VOaRbI1hnz2mzbfscZQcAKix2lrtFapQ9AbKItcTmwV8HNaoU7xPJIvccyy2lmuKJFoP8575ER4+cAF1+74DwJ69FWGt4v9sV7FXSya7tJaZQyKJCfHDYBBMGRzesRopJypGH7j6U9ki7ut7+VPWEuZa17b5lk2Z7vmP7NLaNo5UKHoeJeR60uBhkZsdPSDrTiQhr3avlx1r8ZBFtV8CUKCFUf/xzWCr48BO6XLYY09xHecZtdIviB8Pv1oFt/9MlU84s22bWj20uKqOvXmVzHHExxf3t38LhdehhFxPbHVui9x0IlrkHkJeeqTxvuNb4fPfc3bDt6RHns1z4XcSZMnju/eXcXzPj2RrUZQTRPp9pwFw/qSE3ht3byEEhCZSZwgErfWolQMF8v/0VEeETkm1EnJF36Ice3pis7gtcpMjIeaEEnKPsZRmNt73vwuhtgSTgPSUJcwZMpwD7/ybaXv/SpCwUDjySn5cPJ/IIDN7H1mEuR9XB9SEoU0hL3UI9zBHPfOSalX6VtG3KCHXE0+L/IR0rdTISb36msYWeVWBDJUEVjVMhNBEFoyO48iZjyP2v4s9diix8+9xtW7r7wkwmhCINqJWSh2VHdOi5I91v3MzKbwOJeR60sgidwi5pzujr7FWyx+YwGgoO+re/pPMjiz55Scsfb2MB82yGUPKjPNgxnm9P84+x4CgDYu8Rgp3dLCZsABf5VpR9Dn99/m4L2jkI3cKeWXfjacp1io5rvDB0iLXNCnum1+E4Yu5Jz2EOkwE+Q3s33dNGNt0rZRUWwk0GfHzNRIRaGJndlm/qoet8D50EXIhxCtCiAIhxG49zue12CxgdAi5jxmE8cSzyE2BsuRr3i74x1h4PAFqSyibdBNf7ZFZikHm/u06aRchMLRlkVdbCXc0nogIMLEju5yvVIanog/RyyJ/FVik07m8F0+LXAjpxjiRfOT1NVLIU08Bv1CoyJbRNac/xg91w12HDXTjUhOGNn3kJTVWwgOkkD9+wThAZXoq+hZdnqE1TVsjhEjR41xejaePHKRInlBRK9UQECW7yY86C2pLZXEvYNvyPQDEhfh5d5amHggDtOkjr3dZ5MNjgxkcGUBm8Qn05KUYcPSaj1wIsUQIsUUIsaW/tWVy4WmRg7R+Tyghr3KHRYJLxAGOFFczJiGEjfcsIDLIy2ql6I4Bg2Zv1e9dWVtPiMc8QkpkIEeKlJAr+o5eE3JN017QNG2qpmlTo6Oje+uyvUtTi/xEc604feQtcLS4hpTIlvcNNDSDAYFGg71lIa+xNhDgUd0xNUoKuZrwVPQVKmpFT5pa5AGRUJXfd+PxxGZt5Erx5M8rMsgsqiY5MqAPBnYCIgwYsWNrVcgbF8mKC/Wj2trA9mNlvTRAhaIxSsj1xLP6IcgGwUX7ocHWd2NyUpghG17EjW202VLfwPM/yE5GiWH+fTGyExADBtG6RW6ptzdKiopw+MvP//d6vnc02VAoehO9wg/fBjYAI4QQ2UKIG/Q4r1fRYAOtobGQx4yWNclLDvXduJzk7pTLuAmNNm84LOuOXzptEBdMTuztUZ2YCIdrpQVXia3BjrXB3si1EuGIYAE4WHACudIUAwa9olYu0+M8Xo3NIpeerpWYUXJZkAHRI3p/TJ5k/wTmEMr8kwjz2LwvTyYs3fuLUaqmthODAQN2GlroElRb3wDQWMiDTM2OUyh6E+Va0QtnfW9Pizw4Xi6rC6H0qNsq7m3sdti3gpyoWUx85Dt2ezRFOF5aS6i/L8F+vn0zthMRYcCA1qKPvNYqhdzTtRIZ6BbyR7/IYPmOnJ4fo0LhgRJyvWjJIvdzdJIvy4Knx8PzJ8OKu6Sw9ia526G6gLWGaQCsO1jEliMlHCqsIqeslgTlG2+MMGJoJWqlJYs8PLCxRX7b29uob+jl/2PFgEY9S+tFS0Lu6wdGE+xf6d626TmYeLlsZNDTaBrsfBeOySYJ+/wnA7Xsza3giRV7AVn4aUJSWM+PxYsQwuFaacFHXuOwyP09LPJgc/Ov0b68SsYmhvbcIBUKD5RFrhcu10qTZBq/UHfJ2HOXyWXWht4Z049Pw8c3wZZXIHIouyqk22fjYXebssLKOhLD/Fo7w8DEIeQ7jpVRaalvtMsl5B4WuRDN29z9rFL2Fb2IEnK9cFnkTUTRHCJbwAGMOhtCB/WOkGsabHkZYsfBjJvh5D+Q6cg+zKuwNDp0/qjYnh+PN2GQPvKb39zK79/b0WhXUZX8v/RvUpN9+wML+etF47n9tOEEmX3YnVOOQtFbKNeKXrRlkYOshGgOgdgxULi/58dzPF365s9+BqZcQ4WlnqJ3vmZCUig7sqXIvHfTTJIjAogLVRa5J8Ix2Qmwx2NiuLTayk1vpAM0i/AJCzDxy6mDAPgmI4+skppeGq1CoYRcP1qzyP1CHMtQWRExcigc/l5OeBp0fiCyVEifeIMVVt4jM0tHnwvgqgVyxtg4l5CPTgghqAX/7oDHEUcONPqR255d5lr3N7X+fxdk9nG5YBSK3kB9i/WiPYvcmRofOUSKfmUOhCbpO4Yf/gIbnnW/vvhV8A8DcLlVFoyMJcDXyKh4JeKtYpAp+gBbs8r4ek8ep4+JY3tWmesQ/zZi7gNMPhRUWlrdr1DojfKR60VbPnJwCSqRQ+XSM5JFD+wNkP4qBMbI1wsfgdS5cmgNdl5bfwQhYHBkANfOTmV6WqS+1+9HOKNWnCxxuFN2ebhZ/NpoPh1gMlJTpyxyRe+hTDK9aHD0bWzmWgltvBw0HQbNgC/vhIhUGDJfn+uXHpFlahf/BQbPhrDBrl27jpezNauMRWPi+n3jZF3wcK14kllUTbCfDxMHhREW0Ho2Z6BJuVYUvYuyyPWipThygMQpclmQ4d5/5YcQkQZf3a1fQa2Cn+UyZpT8gfDwvzsn3u44fXhL71Q0QcM92emkvsHOsZIarpoxmDdumI7R0Dzk0Im/yUi19QQolKYYMCgh14uWUvQBRp8HYy+C0x52bzMHwYL7oXAvfHG7PtfPXCuXUc1ruhxzCHlSuCpT2xFqbWDAzoWTk1gwMoZgsw/ZpbXY7BqpUe3XbA80G6mxNqj65IpeQwm5XrRmkRsMcNHLMOGSxttHnwtzfg9bX4cD33Tv2gUZsPl5GDxH/kg04VhJLdHB5kZJLIrWCQs0YxAaF05OZExiKJV1Ng45qhp2RMgDTD402DWsKk1f0UsoIdcLp5AbO9Embd49sofmO1fAphegob7997TE0fVyefbTzXZpmsb+gkoGhat6Kh0lPNCPxFAzs4ZGuVq6OSc6Uzok5PIHU014KnoLJeR6YasDBBg7UUXQ6AtjzpeZnyvudNVE6TQ522R4Y+SQZrtW7slnW1YZC0fHde3cAxEhEJq0pkP85f/nzuwygv18GlU6bI1AR2ii8pMregsl5Hrh7NfZQt2NNpl/H8y7T65bu9DA194AR9ZBwqQWr71idy5RQSaWzE3r/LkHKsIATiF3WOQ7s8tJjQpssa5KUwLM0iKvVZEril5CCbleNO3X2VH8w2DkL+R6fRfSuvd9CaWZMOmqZrss9Q2s2V/I3GHRbUZZKJpgMHoIubTIi6utHW5O7XStVCshV/QSSsj1wmmRdwVfh/+6vrbz783bDQhZkKsJz31/iNKaei6aqnMGaX/H0yL3d7vKOjLRCe46LDV1yrWi6B1UQpBedNUiB/B1hAV2xSKvyofAqGa++ce++JkX12Zy8rAoZg2J6tq4BirCIF1WQGyI+8e5o0LuLH1Q6KiUqFD0NMoi14u+ssirCiCocRlaW4OdF9dmAjA9NaJrYxrICIMsA4xsvOGkIxErACPigokJNvPh1uM9MjyFoilKyPWivlZ2BOoK3RLyfAiMbrTJs4Tq2RMSujamgYyHa8WT1A76yH2NBs6ZkMCGQ0UttotTKPRGCbleWCrc9VQ6i9EXDL5ddK00t8gPOpJXPvnNbAZ3UHwUHjQR8ueumMwvxscTGtDx0NIhMUHUN2g8uHy3EnNFj6OEXC/qKtyVDruCb0DbFnn+Htj7ZeNt+1dCeZb0kXuw7VgZAEOilYh3iSZCvnhcPMsun9ypUwyOlPMe/9uYxebMknaOVii6h5rs1AtLOfiFdf39vv5tC/lzs+TyIY8WYqselcvoka5NFZZ6Xv3xCIvGxBHs14nkJIWbVlwrncEzVLG4Wk16KnoWZZHrhaXC3Q2oK7Qm5JYK2P2h+7XdIzZZ0yBhMky83LVpf14ltfUNXDJtUNfHMtARBtC6FwMe5xHtcstb29h4uLi7o1IoWsW7hPzQalj3j74eRXPsDWCt1MG10oKPfP2/4IPr3a9rPB7Tq/IhfrxMYHFw2NEJKE25VbqODha5wSD4+va5rtd3NGnirFDoiZcJ+SrpTqgq6OuRNKauQi67OtkJrVvkZUflMm2eXFYXyqW9AWqKmk10ZhZV42sUJIapIlldRgchBxgeG+xaL6uxYleTnooewruEfNKVYLfJBsNObNauhe3picXht+4J10p5NiTPhLl3ytdOIa8ukmITFOM6tKTayns/HWNwZCA+Ru/6rz2hcD7h6FBP/K7FIzEIma6vimgpegqv+rb/NV3jZ+NI2PS87K7zxvnwn9nw50FQdLDvBmbRwyJvxbVSegTCU9yx4nk75bIqXy49LPIX1hymuNrK0lOaV0FUdALh+FroYJX/+pQhPHb+OACqVMq+oofwKiFvsGu8Zj0Vyo/Bxn9LV0vRfrDXw08v9d3AnBZ5t3zkLVjklgqoyHEIuSPE8Ov74Lv/g8pc+dpDyD/ZdpwFI2O4cIqqrdItnBUO7foUvXKm7BdXWfnFM2v5bEeOLuftCMdKakg/qsIf+zteJeQRgSberT8ZyzkvwKVvw1UfgzCCKRiyN/fdwIr2yWWTDMtOYQqCukr36/JseO8qaR0OPU2GNjpLAGx+ETLXyPWoYQBUWurJq7BwkkrJ7z46WuQAwY5SuB9uzWZPTgX3fLRLl/N2hJP/upoLn9vQa9dT9A1eJeThgSZAUJhyNow8U3agv78QTroRcnf0na9880sQP1E2Pu4qAeFQ67Cc6irhudlw+HtY9GdImipbxv3xMFz4spxc3fCs/AHzDwcgr1x2KIoL7WKZAIWbHhLyjxy1VzTo9WxP1T+0f+NVQu7szlJcbXVvNBghdqycBC3J7P1BaRqUHIK0UzvfVMKTgEhZeMtaA9vfAksZnLsMpt/kPsYUKK1zp+Vvd/tccxxCnqCiVbqPcE526iPkQWaZmFVeK1v5VdXZyCnrXaND1Ubv33iVkIc7hLzUU8hB+pDBHarXm9TXQoO1exOdAP4Ol0hNMeTuhOB4GaXT7LgwuPMgzL8fLnvbtTnXIQzxyiLvPj1kkQNMHBQGyAijnqbO5hbvZt8ZRb/Cq4S8RYsc3EJeeqRXxwO4Jzr9w7p3noBIuawtgcocCGmnauHcP8CQea6X6UdLEaJx/WxFF9FZyIM8hPzkYXLSuqSmY8L6+c4cXlp7uN3jDhZUMe6hlexw1NkByC2zuNadTwOK/olXCXmrFnlAJPgGQmkfWOSWMrnsrkUe4GGRV+RKi7yD/HSkhPfTs5mQFIavih/vPnoLuckt5HOGSiEvrbay4VAxJ/91FYWVrddiueWtbTz6RUa713jmuwNUWmys2J3n2rYp010WoLSDPxwK78SrvvXBZh9MRgNFTTuvCAERaXD0R2joZcvDlQwU1r3zuFwrJTLksD2L3INXfzxCiJ8Pr143rXtjUEh0FnKDQRBk9iEqyMTIOBmiWlJt5dr/buZYSW0jK7qrZOTKXAan5Z1XbuGZ7w7ia5TzNqU1yiLvz3iVkAshSAz351hpDe9szqLc88M561aZLLNvRe8OSm/XSvqrUFfeYYtc0zTWHijkzHHxhAWYujcGhcSgr5CD9JOPTgglxN8Ho0FQUm2lzibPn1ve/YnPAodVn1kka9G/tuEI+RUWnrtiCiBLBCj6L7oIuRBikRBinxDioBDiLj3O2RrJEQF8uSuPuz7axYPLd7t3ODvRF/dyhmdtmVx22yKXYYQcWSuXHbTIs0trqbDYGJfUTdeOwo3OFjnA3WeO4rcLhiKEIDzAxNFidxbvsdL2hbxp+OC+vEqWrZafdUt9g8sSP1hQha3Bzo5jZYyKD+GUEdEYBKzZX9grE6yKvqHbQi6EMALLgMXAaOAyIcTo7p63NZwF+6HJBI45SFq1vR25opdrxegDV3zgfh2e2qG37cmR1x+boIRcN3pAyM+ZkMCUwdJ9FhHoy968Cte+7NLmpRlyymqp9QgZtNQ3HsuvXt/C31buo7iqzuVjXzAyhqIqKwv/sYb1h4oZlxSKr9HAeZMS+TajgNvf3a7b/ShOLPSwyE8CDmqadljTNCvwDnCuDudtkeQIt5AHNW2cEDa49yc8XZOd3UjPdxI7xr3uyNhsj61ZZZiMBkbEBbd/sKJjOIVcpxT9psSG+HGoUJYbNhkN7M2rbGRx7z5ezqwnVvHs6gOubTVNCm75GKTv+2BBFYWOOaMrZwzm/EmJZDpKGU9MCgPgwbPl5+qH/YUseX2LSg7qh+gh5InAMY/X2Y5tPYJnadAgs7HxzvDBvW+Rlx6RE5VGHbrxBMW5151RLO2w8XAxE5PD8PM1tn+womN0xiJ/7xr4eGmnTj8kOsi1fuWMwRwurGa5R/2VH/bLCpdrDxS5ttU0SehxZvAeLKxiwyEZnRIdbOaJC8e5jpk5RM67hPr78sdFIwD4+ud89uRUoOhf9NpkpxBiiRBiixBiS2FhYZfPMyk5zLVutTWxLMKSZY2S3rI4NE0W7ko9uc3DXlxzmPWHito8BnBPsnWQGquN3cfLmaHqq+hLRzM7LRXw8yew4y14aSFYqzt0es+mHzeenMrgyADu/GAnlRbpKlzjEHJPn3ZTIQ9xPI3uyi7nbytlrZ/EMH/MPu4f9EEeT6+pHq3nvtyV26FxKrwHPYT8OODZVyzJsa0Rmqa9oGnaVE3TpkZHd724lGcfyo2HixunOgcnyCzLml6q9lZ8SFYhTJvX4u7X1h/h4c/28NiXGVz7yk8dO+fNG+HWrR069GhxDXYNhiu3ir501CL/4g73evZmWP9sh06fFuW2yONC/Lhr0UisNjtHimr438ajbHI0a872mARtWsvc+fqLnVKU71g43JVnsfJ3c/n0N7MbHZ8S5Rbyl9Zm6hIpozhx0EPIfwKGCSFShRAm4FJguQ7nbZXlt8zGxyA4XlbLrCdWuUOrgh2uicpeKhNa4si48yiWtWJXLmsPFHKspIYHl+/hvz8ekUPz62Cf65hRENmxeuLOyIfBEaqtm664hLyNJztrDex6T1atdLLx340rWLbChEGhnDwsir9eNB6DQTAkRp5jX34lDy3fw4SkUOYOb2zs1NQ1tsirHbXNKx3L+aPcDUZGxAUzwVEKwMmo+BA+/c1s3vrVdKwNdg4WVLU7ToX30EF1aR1N02xCiFuAlYAReEXTtD3dHlkbjE8KY1pKBBscDW335FQwe2iUO2SvMg/ixrVxBp1w+uPDBgOwLauUpW+2bE3beqDa3bESKeSeE8AKHXAWP2vLIq9wPHQuegKyNkLKHPjk15DxWaNm2C0R7OfLGzdMd71OjghACPmEabNrXHZSMj/nVrDG4z1NJzurmwh7SmT7P+YTBoVxqFAKeLOkOoVXo4uPXNO0LzVNG65p2hBN0x7T45ztUWFxhx66Oq84k2gqeskiLzsKRrOrucNXHunRTnY+dDp3Lx5JeW09O46V8d8fM1mhk4/ySHE1of6+hAboMNGqcOOyyNuIWil3zO9HDoHzlsGES+Uczc734Oh6WTO+g2WV/XyNDAoP4JNt8schNtSP+SOlhe3MzPT0kd/42hb25VdymocVHmjumE0WFWQGZJMLRf+h2xZ5X1HmkdXpslZcrpVemswpy4KwQa5Jyv35jR+ro4JMhPj5uiadzl32o2vfkSd+0a1LHy6s4v0t2a4iTAodcQp5fQ1UFUJQC3M65dlyGeII0BICJl0Nqx+Fw6vlNh8zTL66Q5e8YU4qDy6XD7LxoX6MjAvhkXPHEB3sx6//l84P+ws5b1IimqbxbUa+4zh/vv39KZ1K9Anxk2UuCpVF3q/wqhR9TzyTgVyPmUZfiBgiLSLosThgQCYCHV0PUSNcmw4UVHHOhAQ237OAf1wygQ9+PQuAqYPDdb/8npwKrA12/nDGiPYPVnQOZ/PlF+fDk0NleGFtaeNjyo8DonEG7oxfQ8JkGHEmGE0yoqmDXDzV3Z4vzlHB8qqZKcwdLn+oP952nMLKOk7/h9vh4mMUDI0J6lRXKCEEkUEmiiqVRd6f8Fohv3X+UNd6tWdT23EXQ+YP8N9fwN+GwJEfW3h3Eypy4fsnOldwK/012dF+roxcqLHayC6tZVhMEDEhfpw/KckVKRDTQmnZP69ov6JdWzir2TkflRU6Ipp8LXa8JT8fTjRNttoLT2mcP2AOhiWrZZ34sRdC5toOXzLAo0JiqL9vo+03zU0D4MW1hzngMUnZ1RrjUUFm5SPvZ3itkN90yhAOP34mQjTpfjL1eukrP7pOWlHf3N/+yd65HL7/M+z9omMXz9slz5swCRJlUaIcR+3n5MiWJx6/u+MUYoLdovv8D4exd2MCtLRa/uiEKf+4/ngK+fnPw8Qr5A+3M048ZytkrYcZN7d+jqhhUFPU4dhygL9fPIFrZg5GNOk05fSXv7Upi+hgM1fPlJPrzeryd5CoIBP5FZb2D1R4DV4r5CDLgwb4Ghtb5MGxcON3cNNaWPgIHE9vtQXcwqd+4Devb5RfTJATVR3h6/vkctJVrk2unpmtNHYYEh3Ed3ecwhXTk13bulMjurTGSrDZR9Uf7wk8hdw/AiZcBrZaOPit3HbgW0DAuItaP0eoI7WivFlKRatcOCWJh88d22x7lMMAqKqzMTMtkjtOH0Govy+/PqVjYapNGZ8Uxr78SlVEqx/h9SoQaPZpFppFaCLEj4fUufJ1Xstdyw8UVLH9573uDfm7WzzORdFBWHmvbIo8fSlMu8G1y5lgER/aes/MYD9f7j9rtMsdktcNq6isxkpYoLLGewRPIQ+IgOSZYA6V/+8AB1bKp7G2yig4J0HLj7V+TAeJ9niSG58USqi/LzsePF2G3HaBBaNi0DR3BqnC++kXQl5V18qkpmdceSvECUcWaOJU+aWztWGlbHlZdq8HGLG40S6nRR4T0rbP2s/XyAtXS3dMQUXX/ZQlNfWEq/rjPUMjizxcVqYMHywjVTLXyqe88Ze0fY5Qx+RlRcct8tYI9ggtnJrS/XIMzuYWLVVdVHgnXi/kASYjNXW2VnZGgcGnxXBEp3863iHktQkzZAJIWVbrFyvIkOVlb94Iaac02pVbYSEy0NSh4lXOvprd8VOW1VhVI4meoqmQg7SwK3Lgh7/IvIEp17R9DqcRsfxWqO5AnZ22huPhMx+f2MlyxZoGqx+Hz37nmnw1+Rgw+xiotLTyvVF4HV4v5NIib/yB1DSN//6YSWF1vawo6GmRr3oUnhpNTZEUbKdFnhUirWRX2n1LFGRA8oxGKfmuXRWWFqNTWiLa4VrZeLi4Wez55sySFlt/NS09WlpjJVxNdPYMnkLurDMfkiBdb0fWwoyl4Nu6Cw2Q0Sxzbpfrqx7t9pDuWDicR88bi8Eg2j/Yk+KD8scn/b/w2lmQuwMabAT7+VKhhLzf4P1CbjI2qwx3rKSWhz/7mZvfTJdJQk6LXNNgzd+g4jjWrHQEdqYb9lKtmdmtORo5tCbkFblQldeiiIOsVBfRQZ+1ycdAbIiZT7bnNIoLBvjl8xsaJQ6B7KQ+6ZFvXKK/J6ecYyW1rkdkhc54CrmzImWoR2Vm59xLe5z2kJxLSf8v7P+6W0O6dcEwrpwxuHNv2r8SPv2NXD/lLllU7vm58EgkI02FjbKjFd6N9wt5CxZ5Tb18fbCgqrGQO5tAALbiTK4xfs1CYzovNfyC/VVmMIe0LuRbXgYEjDyrxd1lNfWdcnV41sbYfby8zWNveWsbZTX1ruPe/ekY/r5GLveIgFHoiNlRCCvVw30W4iHksc0jS1pl4f/JsrjZHax+qReWcnj7Uji2CeLGw7y74cZvIOkkACYbDirXSj/C64U8PMDULIzP2ZS5vLZe+jOrCuSOCg9feWkmVxu/pixmGm/6XU5ZjQ0iUlsX8oPfysJIrVQm7KyrwzPp48Ln1lPfYG9kIbXUxcUZ5bL+UDEnpUY0OodCR+InwtL1cNUn7m1p8+RnKW2eTL3vKD4mOfHZlstOL9Jfg7+mwbMnwZMj5JzPZe/ArxwZpqFJcO0XIIykieOu+ucK78frhTwi0ERZTT22BnelOqfvz64BgVEyMcje0GjS06f0EMmiAEvcSYQHmimrtUJEWstfOLsdCvZKy6YF7HaN8tp6IjphkdsdQh0VZKbOZie7tJbcMvfkZ3ltPVabvdHTRn65hc925HCwoMrV/UXRAwgh2+55NvoIjoU79sGVH3X+fK19rjpKTUnHyk1kfCbdQsGxMu49eSYMX9Q4+9THBBGpLKj+EmNNJyZh7Q3w/V+grPvhlAr98XohjwyS4vltRr6rRKdnHRa7fySgodWUsGuvIy1+8Gwi8tfjI+yIyDRCA3wpramXdVrKjsK+r+SXonC/7AFamim/GK34xyss9dg1OuVaueP0EYxNDOFvF8kfh8yiqkZNMnLLLfzpw52MfXBlo23LVh8kNsTMpdMGNTunoocRotNdnAAp5KUtJ6W1S10l/DUVvnmg7eM0DXK3w9CFcNWncN5z8Ms33CV5PYmfQLC9nOerbgNrNTuzyzjtqR9cT7ItkvkDfP84rLy7a/fRDpqm8a/vDjSb/Fd0DO8X8kD5mPvr/23l9ne3Y6lv4KOt2a79FlMYAN+m72Hlhu0A2BY/6drvGz2UMH9f+SEeeSbYbfD2JfDulbBsGjw9Hj68UR7s2RzZg1V7pesmvBMJOqPiQ/j81pOZ6GgAcLiwmiPF7nTu46W1fLytcQzy1z/nszevknMmJKjQQ28icoh8KuxKGKLTLbjr/baPK82UtX8SJsofm4mXt1y1EeCsf7I2/loiKYOCDJ78ej8HC6pc9f0b7BrrDhRJgyh7i7TG962Q763rmYYUxdVW/v7Nfm54rZfnEvoJXi/kEYFuQcsrt3DZixtZ72hGC2DxlXHAm3cfYLThKBXmeEoC0si0yxriwYkjCA8wSddKwmQYdkbjC/gGQOFemH+fzOZrgd+/twPonEXuJDzQRFiAL+lHS9l0uIRgsw8GAa9vbNxEeqJHx5dUj1ZhCi/A+SRX8LMU9M0vQl47WcROXOLfgmVdmQ9vXSrDa1c9Cj7+0pXSHn4hHEo4DwBb3h58HCGNzpT9x7/M4MqXN/HoM8vgpQXw5sWw5b/yvcWHOjbuTuLsdtVb7Xb7G15bj9yJ07UCUFBZR0Fl42zJap8wIgBDbRHTDRlkBp1McJ2NC60P8/ycak4KjScsoIzSmno0QFzxnqzN8sxEGanwu92ywYCxZWvb04fd1cnHK6Yns2y1/IJcNCWJw4VVrNlfSJBHRM6TF4/ntKdkqGJKK4W5FCcoMaPlsiBD1vPZ9oYsc3vp27DzXRgyr/WuQtUOi7wlF8mOt2H/Cvg+Do5ugNHnygzUDhCWOJSaLWa2bVxLlc9wQLr3AFbukXkXJ1d9JXt+HfoOAmNgwiWw/l/yx8iZKKUTWSXyaVRN4HeNfmWRt0S1TxgAd1c9QaSo5GfTeKrrGiglhPK0swEIDfDFarNjqXdMmIanwOK/yRl+g6FVEQd3av7c4dFMatInsaPcMm+Ya33h6FgumTaImGAzD5w9mrMnyAzBhDB/nr50IgAj41X8uFcRFCuLb634oxTxtFNlk/Cv75V9Pz9Z2nwy01oD29/2CFtsQcidrQaPrJV9aiOHNj+mFc6akMgh0wiSC74n/Ugh5xrWYVn/PG9uPEJ2aS23jqjgLMNGjqVcjGX2H2Vp3rRT5ZtbqV3UHZwWeYd72yoa4fX/ahEBJs4YE8ug8ABeWtd8QqmCxh3m14lJpDis3CBHDQtnzZKyWiv+Jn9p/Uxf0qHrFzhCAn99Slqz8qMdxd9kZNGYOL7fX8CCkTH4GA1cMk3GiJ83MZHfLxxOgMmHcycmcs6EhC5fR9FHCAHnPCOjPmy1chLyqdHSZedk/TPuTFCAj2+CjDZ6mGuarPkCMnsTZPhsB/ExGhhzwV0Y3r2c23w+5rc+Mhrnsy/28gefWG45+ikFhLF47xlU7Q3gyMKpbn997s6OJ0W1wuHCKl5Yc5iHzhmDn6+RndkyR6K23t7OOxUt4fVCbjAInr9qKgWVFj7cms29vxhNWnQgGw4V87eV+6huMMJVH8Mb53PIHs/h2kCXu8Ip5GGOx7nS6vo2qxe2hDO2u7XytR3lmcsmYbPb8WlSltbkYyA1yp08pETcSxl1tkwms9scnaxSIW+no4bLcfj2Iagtg3n3yJouTUXc4pE0dmwzvLxQrg9fLN0r0GqOQ2sYRp5J/Yiz+e0+KeIHAqdwdvVGuTNiCPfmnUsV0o1XVmMlLChG1vrP2gCzbmnxnH9buZfEsIA2k9UKK+uY//cfAFm6V9PcAQMVtSq2vSt4vWvFSUywH1vvX8hFU5KYnBzOwtFyMrPaaqMhdR7jLS9yrvURMnIr2JldBkCgWRa4cjYvLqvtfH1mp5DHdlPITT6GRl1iFP0QIdxuOmd1xMGzZXx63Dj48Z9waDWUOCYUZ//O/d76ahlBAvCVIwQweiScu0y6boITIGp4p8fje/4yOYb59xO05Au+m/ws5VesgNu2UpPmnvh/ZV0mlvoGSoecJyNYWqmz/sm2HD5IbzvW/IkV7ieRYyU1/OH9HaRGBXLuxATKulGjfyDTr5TD01oNMEmRrrE2UGO1UUEgCaF+VJVb+Ncq+Sga5NfYtdJmHG0rHC+tJcTPp8NdzBUKAIYthONbZZ/P4DiZaPTkMFl909kz9KRfySiU+mr45Dfw/nVw47eyEcopf5LWO8DvM2QiUFee1vxC4bovAYgH4s9xN0t56epp7M2r4Px/r+eZVQd5ZtVBpogYPjQ3yAgcz/ozDspqrFTU1qNpWotPj7XWBr7anctZ4+P5YlcuGw4Vk1VSw8PnjKGwso7y2nrsdq3zxcEGOP3GIm9KoMO6ramzuZoz/2b+UCYlh7mOcblWHBZ5aQeFXNM0rDY7V728iTc3ZZEWrcIBFZ1k6vXwh32uVoEERIHRDOVZchLTaJJW9uCZMPQ0uPBFue/vw2XqfcrJ7nMZjF0T8XbwNxmbFWarwOHmq2ueuGO12am2NlBZZ2u1acrSN9OptjZwweRE4kP8eD9d5nxMS4kgLMAXuwZVTRvFKNql35qR/g6LvNra0MgnPjIumG1ZZRgE+Dtqh4f5uyc7O8ITK/by/Bp3ynWahw9boegSBoN0t5Qdkw0sQgc1ziJNnQtn/Fl2Jxp6mnSH9AL+JiOXnZTMkOhAPt+ZS8ExR1x7C0Lu+f05kF/VbL7JUt/AjweLOG9iAvNHxpISlUlOuYUh0YGMiAtmX34FIBuuhPipMMTO0G+F3OxjwGgQ1Fhtrp6egSYfBjuqDto1tyvG32TE7GPosGulacalrRtNlBUKF6FJ8PMncn3ilc33z7xZ/vUyf75gHAA3npzGwseLwQpYm2d4en5/9udXMne4O7PUUt/Anpxy6hs0Fo+LB+Chc8awZn8hZ09IwGgQJEfIidVjpTUMjVFPuZ2h3wq5EIIAk5Eqi40fHL0JA80+jSJAPAkL8O1wM+TUqEBqrQ28vWQGZ/1rHZedpMrJKnQgPEXWNDH4wNlP9/VoWmTOmBTYBpaqMvyQ6fxGhz+7zCPi5GCBW+h3HCtz1dj3MQgmJ8tkouGxwQyPdYcHD3IKeYlqQddZ+q2PHKQF/tqGozz1zX5AulZOGd5y/Ykwf1lFsSNUWGxMT4tkbGIoR574hapEqNCHoae5140npo11/pRkajQzmcfz+DmngvEPrWSDoySG8/sTbPZpVPxqX55cDzAZee7KKY2aSXvi7Jz1wKd7VPGsTnJiflp0IibE3GjSxd9kwM/XyPJbZlPVpKh+Zyzyitp6RsUHt3+gQtEZnEI++7d9O442GJcYSokhgOz8Al75/GeqrQ3sOl7GzCGRrtDBKSnhpB8tdUWu5Dqyn7c9sBCzT+s9bT2jXL7LKGhkrSvapl9b5M5HuCmDw5k4KIzEMPnoNj4pjFlDoxodGx1sprBJnZbWqLDUq8kYhf6YAuDePJh3X1+PpFWEEBj9grFUlbmqJWaXyvLLxY6iW9NSIqi02MivkN+n3PJaooLMbYq4k9eulx2MutOYfCDSry3ypacO4XBRNX+5cFy7GZuxIX6s2lvgsiI+25HD9LQIYoJlos8H6dnUWm1cPn0wlRabKu6j6Bnaa+p8AhAUEk5obR0nJUVQXlPvEvKM3AriQ/1cBtSBgkriQv3IKbcQH9qxhLlThkczJiGkUUlnRfv0a4s8NsSP168/qUNp9zHBZmocoYql1VZufXsbS15Pd+3/w/s7uP/TPS6XTIgScsUAxcc/hFlJZt5dMoPBkQFkl8rJyV3HyxmTEMqwWBlxsienAk3TOF5a02EhB9nP1llES9Ex+rWQd4aYEDnRkltucfnVM4ukVVBa7fadH3WU2wxRVdoUAxVzMD711QghGBQRQFZJDXd/tJPDhdWMSwwlKsjMhEFhLFt9kD9+sJNDhdVMS4no8OnHJIaQWVTN1qxSCiotfL4zpwdvpn+ghNxBrMOFcvo/1rjixJ1zL1uzSl3HbTpcAiiLXDGAMYfI/rcN9SwYFYOl3s7bm49x/qREV7GsZZdPIjLQxPvp2QyJDuSGOR2vzHj1zBRC/Hx4d/Mxbn93O7e8tY3DhT3Tmai/oITcQYxH0asX1jRulPv9vkLX+qZMKeRhSsgVA5Ux50FNEez6gBmpkcwZGsVNc9N46pcTXKGFSeEBPHa+TCQ6a3xCp2qnBJl9GJMQyp7ccnYck1UfV3t8B1vDarOzam9+5++nH6CE3MGQ6EDuP2s0Uwe7O59oGny9J483Nh5l4ehYAkxG0o9KIW8tFlah6PcMXyRrwRTuxWAQ/O/G6dx95qhmRbJmD43io5tnccv8jje8cDIsNojdxytc5TVW7snjrg93umLWW+Lej3dx/atb2JtX0enreTtKyB0IIbhhTioz0tzJPRWWej5wFPW5bf4wBoUHuAprxXSzbK1C4bUIId0rde0L5uTkcHyNnZcZZ3IQwFnj49mcWcI7Px3j+ld/otLScuLeh46m65WWgVd0Swl5E1I8Uvg1Db7fX8hpo2IYlxTKoAgZ/RJgMroqJyoUAxK/kMbNLnTmoqlJnDsxge0PLOR6h389MtBEbX0Dd76/s9nxmqbhLHnUlXLU3o4S8iacNiqGa2YOZvM9C0gK98dqszPEUcDHWQtCuVUUAx6/ULD0nAsjPtSfpy+dRFiAicnJ4Wy57zS23Hcai8bEsTGzmHUHilxhjwCVHk3QywdglyEl5E0ICzDx8LljiQnxY+mpsnXWsBiZKjw2IRSALFXURzHQ6aBrRS+igswIIZg1NJKymnqufHkTV728GU2TZniRR1a2EvJOIoS4WAixRwhhF0JM1WtQJwq/nDqIR84by5nj4gBYNFYunaGKCsWApYddK60xKt7d6CKzqJo/O9rGFXvkepQNQCHvrqN3N3AB8LwOYznh8DUauGrGYNfrQLMPHy6dSVSQcq0oBjg97FppjQlJYVw3O4VrZ6XwzHcHeWHNYeYMjXL1HICB2cC5W0KuaVoGDKzO7lMGdzxDTaHot5hDpUVeVQA522HI/M6X3k1/DUoOy96j+1dC0X4YcSagQeyYFt9i8jHw4Nly35K5aZRuX86tr1Rx9fyJAPj5Glp2rWia7LwUHOdugN2P6LXQCyHEEmAJQHKyasSgUHg1fiGyKfSTw+Tr+ffD3D90/P0NNvjsNrlekCFb2AGsekQuL3oFxl7oPj5nG+z+EAJjZGEx3wBGRI/gFdOTHLLHc8aqvwA+pEQGNhPyR95YwdUFTzC4cjvMvAXOeKxLt3wi066QCyG+BeJa2HWvpmmfdvRCmqa9ALwAMHXqVNUbTaHwZqoKGr/Oax4S2CZH17nXnSIOED8Rakvg01sgO11a6+Yg+PJOyP6pxVMNMeTysM9rPGn6NbEhfhx3VGN0ctaB+4gQx0EAO9+F0x4+YRt3dJV270bTtNPaO0ahUAwwxv8SDnwD16+Aj38NlR6p8XY72GrB1EJbRXsD7P1CCqpvAPzi77Dx33DG47KhtMEIFTnw+e2wcRlYK2HWb6WIn/4oTLhc9gstyIC1f4fUk8nam8755TtYeMvJfLo9l8e+zCCruIZ1B4vYcfAYj4tDPNtwPr+97Bx4/1pp3Q+a1mv/VL1B//pZUigUvUPyDLh9l1wPioHjW+W6tQZeOV36z2/ZAj6OwICsjbD6cdmT1Mno82Di5fLPk5AEuPxdeO9qyFwLaafK7WnzIDBS/oUPhhGL5FDCXoXPvifAms2isYk89mUG3+3N5+HP9vCJ6X6MBo10+zCqYyYTCJC7XQm5J0KI84F/AdHAF0KI7ZqmnaHLyBQKhXcQFAdV+XJC8eC3kOcQ+EdjYPylMPV6+N9F0roGGHYGpM6Fkb9o+7whSdLqLz4kX0ektXzckPkgjPDiAgYFxzE99Ld8vO040ZQx0SAL4KXbh5NtC2dEQKScnO1ndCuOXNO0jzVNS9I0zaxpWqwScYViABIUA/U10uWRtRGMZph6A8SNgz0fSwvdWgljL4I/ZsIV78GsWyCindK2wbHyvDnbpaibAlo+LixZnq+uHIr28YRhGYey80gW0o9/i7iHavzJKq2FpJNg+5uQtUnff4M+RmV2KhSK7hHsiIV4/TzIWg9J0+Csp+DX6+C2bXJfaDJc8AIEdCJ8N8hx3gNfQ+SQto897WG4eSOcu4zU2j08MTSDB+dI4X9yybmYjAZ+OlIix+XjB9v/17l7PMFRQq5QKLpHyhy5PL5FTiQmz3DvC02UYr70RzmR2RmCYuTSXg/jLmr7WCEgZhRMvAJMQZydUMn4gFJA4BedypTB4azZXyj978NOg4OrpCuoKZoGP70kJ1y9CCXkCoWie4QmwZ2H3K8Hz2y8PyJNxp13lmCPqOfJV3fsPUJA5FAoPgAlhyAkEXzMLBgVw968SlmrfMgCqMiWCUhNydsFX9wBy2/r/Hj7ECXkCoWi+wRGwYzfyPWkk/Q5Z2iSXC54oHPvixoGh1bBnk8gcTIAF05OwmQ08PHW4zB0AQCZ7/5JhkM6sdXBp457qGm9gQXb34Z/TYXcTsbO9yBKyBUKhT6c8Rjcnd0167slzMFwXyGcfEfn3pcwSS7t9ZAsnw7CA02Mig/m+TWHufmLQrK1KFKLVkPmGvf79q1wJzbVlrR87oocWH6LtPi/fahz4+pBlJArFAp9EEKKr574mDr/nulL4ZdvQNQIGHmma7OzcuKXu/K4wnoPABU5Hu6V4oNyOeM3UHoEtjWZEK2rhI+WyPWUkyF/d+fH1kOohCCFQtG/MBhg9Dnyz4PQAFks64JJidTWRWM9ZKT4+EGczw/lx/dh9ovh24hrOCttDyy/Vf4YOJOHVt4LR3+U2ajWGjiyFqqLZYJSH6MscoVCMSC4cU4a185K4dHzx/LYhRPJ0aLQSrMA2Spuf8Z2dtREcMtHB9l76n/AL0yWAQCoLZXx59NulAlOMaPk9oI9fXMzTVBCrlAoBgTRwWYeOmcMASYfwgN8yRPR+FYdAyC7pIZUkUemXUbKfHeoWpYOOPitnATdvxLsNpmpChA6SC49a8z0IUrIFQrFgEMIQYk5kfDaLNA0MjMPEiUqKA4ajtEgyCu3yMQmez3k74Fjm2QzDedEqnNCt673uyS1hBJyhUIxICkKGU2QvZK6wkNUZG4B4LqLz2NodBB5FRZX6CJZG6FwH0SPRBOCVXvzKW1wtHvsgy5JLaGEXKFQDEgGj58LgPnfUzhrz+9pwEDAoInEhJgpqLBI90n8BPjuYTj6I1rUCJ76Zj/Xv7qFlzbmgsGnVxtQt4UScoVCMSA5ZfZcSkS46/W7kb8BUyBxIX7SIhcCrvgAzZHK/+jmBv61SoYoltbawByC3VJB+tFWYs57ESXkCoViYGL0YdnI17it4fek1b1J3oirAIgN8SO/oo6l/0sn3x7C4V+8zX9sZ7G8YRYA4QG+FFbWgV8Iew4f48LnNvDO5izu+2SXS/R7GxVHrlAoBiyjhg7h5W1VAEwaLK3zoTFBAKzYncegiABmDhnHEzZ384uxiaHS9WIOoTCnEIC7PpI12G+ZN4y4UL/evAVACblCoRjAnD8pkZ9zKggwGTl1eDQAc4ZFufYXV1kprrI2ek9siB8H8qvQEkIIorDRvqPF1UrIFQqFojcxGgQPnD260baoIDNjEkLYk1PB7uPlDIsNarQ/NsRMYVUd9T7BBJOFGSsG7NTiR3Z+EdOrV0PsGHfSUC+ghFyhUCia8MVtJ/PUN/t55rsDrNpbgJ+vgctOSmZCUhh1tgYa7BrHanwYZchim/kmdmspPFJ/FXPX3Ae1h2XzilvT3RUcexg12alQKBQtMCEpFIDNmSVY6u08ePYYzpuUyDkTEokL8WNNth2AAFHHSYZ9fGa+D5/aQipn/glsFtj7Za+NVQm5QqFQtMA4h5A3xd9kZHpaBB/Uz2q03WIM5nTLX1gdd61sbnHw214YpUS5VhQKhaIFYoL9uOmUNMIDTJyU2rjX6LCYID7VUrmv/jpKwiewv9jKZ/ddTtGja8gsrJY+8vzeK6ilhFyhUCha4e7FLU9YDo2Rddf/17CQjNsWYbPb8ffzJSHUn8yiKogYLBtV2O2yrK6T/J8hdnSL5+wOyrWiUCgUnWTy4DBSIgP406KR+JuMBPvJWudp0YEcLqqG8BRosEJlrvtNez6B52bKpc4oi1yhUCg6SUywH9/fOa/Z9uGxwby56Si20GQprsUHITRR7vzmAYifCMMX6T4eZZErFAqFTkxKDsNSb2e/lgwGX/joV/DTS1BTAmVHYcx54Kt/wpAScoVCodCJyckyzf+nYjNc8gZU5cMXd8Cm/8gD4sb1yHWVkCsUCoVOJIT5E2z2IbOoGkYshjOflDu2viGXceN75LpKyBUKhUJHEsL8ySmrlS+m3QgBkVCZA7HjICimR66phFyhUCh0JCHMj4MFVdRaG2RN89ixcsfIX/TYNZWQKxQKhY7Eh/lzuKiai59fLzcs/guc/QzMuqXHrqnCDxUKhUJHLPUNAOw+7mgDFzOqxyshKotcoVAodOTaWSmudVuDvVeuqYRcoVAodGR8UhhPXCDDDHPLLb1yTSXkCoVCoTPJEQEAHCup6ZXrKSFXKBQKnUkM9wfguDMMsYdRQq5QKBQ6ExVkBqC42trOkfqghFyhUCh0JtDsg7+vkaLKul65nhJyhUKh6AEig0zKIlcoFApvJirITFGVF1jkQoi/CSH2CiF2CiE+FkKE6TQuhUKh8GqigkwUVXmHRf4NMFbTtPHAfuDu7g9JoVAovB+vscg1Tfta0zSb4+VGIKn7Q1IoFArvZ0RcMIWVdazck9fj19LTR349sKK1nUKIJUKILUKILYWFhTpeVqFQKE48rpg+mDEJIdz90S7Ka+p79FrtCrkQ4lshxO4W/s71OOZewAa82dp5NE17QdO0qZqmTY2OjtZn9AqFQnGCYvIx8MBZoymptrLhcDHf/JzPbW9v42BBle7Xarf6oaZpp7W1XwhxLXAWsEDTNE2ncSkUCoXXMzE5DJPRwLZjpXyxM5fs0lounTaIoTFBul6nW2VshRCLgD8Cp2ia1jtFBRQKhcJLMPsYGZ0QwpYjpeRXWFh66hBmDY3S/Trd9ZE/CwQD3wghtgsh/qPDmBQKhaLfcPKwKNKPllLfoBEX4tcj1+iWRa5p2lC9BqJQKBT9kTPHxfOvVQcBiAvtGSFXmZ0KhULRg4yMC3atxyshVygUCu9DCEGovy/AielaUSgUCkX7fLh0Fp/tyCE62Nwj51dCrlAoFD3M0Jggbl84vMfOr1wrCoVC4eUoIVcoFAovRwm5QqFQeDlKyBUKhcLLUUKuUCgUXo4ScoVCofBylJArFAqFl6OEXKFQKLwc0RclxIUQhcDRLr49CijScTjegrrvgcNAvGdQ990RBmua1qwzT58IeXcQQmzRNG1qX4+jt1H3PXAYiPcM6r67cw7lWlEoFAovRwm5QqFQeDneKOQv9PUA+gh13wOHgXjPoO67y3idj1yhUCgUjfFGi1yhUCgUHighVygUCi/Hq4RcCLFICLFPCHFQCHFXX49HT4QQrwghCoQQuz22RQghvhFCHHAswx3bhRDiGce/w04hxOS+G3nXEUIMEkKsFkL8LITYI4T4rWN7f79vPyHEZiHEDsd9P+zYniqE2OS4v3eFECbHdrPj9UHH/pQ+vYFuIIQwCiG2CSE+d7weCPd8RAixSwixXQixxbFN18+41wi5EMIILAMWA6OBy4QQo/t2VLryKrCoyba7gO80TRsGfOd4DfLfYJjjbwnwXC+NUW9swB2apo0GZgC/cfyf9vf7rgPma5o2AZgILBJCzAD+AvxD07ShQClwg+P4G4BSx/Z/OI7zVn4LZHi8Hgj3DDBP07SJHvHi+n7GNU3zij9gJrDS4/XdwN19PS6d7zEF2O3xeh8Q71iPB/Y51p8HLmvpOG/+Az4FFg6k+wYCgK3AdGR2n49ju+vzDqwEZjrWfRzHib4eexfuNckhWvOBzwHR3+/ZMf4jQFSTbbp+xr3GIgcSgWMer7Md2/ozsZqm5TrW84BYx3q/+7dwPDpPAjYxAO7b4WLYDhQA3wCHgDJN02yOQzzvzXXfjv3lQGSvDlgf/gn8EbA7XkfS/+8ZQAO+FkKkCyGWOLbp+hlXzZe9BE3TNCFEv4wVFUIEAR8Cv9M0rUII4drXX+9b07QGYKIQIgz4GBjZtyPqWYQQZwEFmqalCyFO7ePh9DZzNE07LoSIAb4RQuz13KnHZ9ybLPLjwCCP10mObf2ZfCFEPIBjWeDY3m/+LYQQvkgRf1PTtI8cm/v9fTvRNK0MWI10K4QJIZzGlee9ue7bsT8UKO7dkXab2cA5QogjwDtI98rT9O97BkDTtOOOZQHyR/skdP6Me5OQ/wQMc8xym4BLgeV9PKaeZjlwjWP9GqQP2bn9ascM9wyg3OMxzWsQ0vR+GcjQNO0pj139/b6jHZY4Qgh/5LxABlLQL3Ic1vS+nf8eFwGrNIcD1VvQNO1uTdOSNE1LQX53V2madgX9+J4BhBCBQohg5zpwOrAbvT/jfT0R0MlJgzOB/Uh/4r19PR6d7+1tIBeoR/rFbkD6BL8DDgDfAhGOYwUygucQsAuY2tfj7+I9z0H6D3cC2x1/Zw6A+x4PbHPc927gAcf2NGAzcBB4HzA7tvs5Xh907E/r63vo5v2fCnw+EO7ZcX87HH97nLql92dcpegrFAqFl+NNrhWFQqFQtIAScoVCofBylJArFAqFl6OEXKFQKLwcJeQKhULh5SghVygUCi9HCblCoVB4Of8P6xj+jzsGSmsAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(range(len(y1)),y1, label=\"qqq\")\n",
    "plt.plot(range(len(y2)),y2, label=\"btc\")\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# EDA \n",
    "\n",
    "We used the yfinance library, so there wasn’t a need to do much data cleaning for EDA. The downside was that the minute data is limited to only a short period of time for download. \n",
    "\n",
    "We looked at BTC and QQQ 5 day price histories to see if they were correlated. They indeed looked correlated. We then normalized the closing prices and performed cross correlation and found BTC to be ahead of QQQ by a number of minutes. We then applied that shift to the price chart."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0, 4657.263934995171)\n"
     ]
    }
   ],
   "source": [
    "def max_offset(df1, df2):\n",
    "#     print(len(df1), len(df2))\n",
    "    corrs = signal.correlate(df1, df2, mode='full')\n",
    "    return np.argmax(corrs) - len(df1) + 1, corrs[np.argmax(corrs)]\n",
    "print(max_offset(y1,y2))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "global_period = \"1d\"\n",
    "global_interval = \"1m\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEWCAYAAABv+EDhAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAAB5OUlEQVR4nO2dd5wbxdmAn1G73t17tzEYjDEG01voHUIPGAgECAmQUEMKCYSaQJIvEEoAQ+glpnewCcWACwbce/fZ5/P1olOZ74/Z1a7q6YrupLt5fr87rbZIs9rdeed95y1CSolGo9Foeh+O7m6ARqPRaLoHLQA0Go2ml6IFgEaj0fRStADQaDSaXooWABqNRtNL0QJAo9FoeilaAPQwhBBzhBA/NZbPF0J80MmfP0IIIYUQrk76vIOFECs647PSASHEbUKIZ7rhe2cIIT5PwefOFEK0CCHWd+Az6oUQozqxWZ1CZ9/LMT5/jfHbdfn9kCxaALQRIcR6IcQOIUSebd1PhRBzurFZMZFSPiulPLq725EIKeVnUsrx3d2OdEIIcZgQYnN3t8PGvVLKEe09WEqZL6Vc29FGpErIJfndbb4mUsrRwJ0palKnoAVA+3AC13T0Q4Si116DVI28uopMb79G02s7nw5yH3C9EKI41kYhxAFCiHlCiBrj9QDbtjlCiD8LIb4AGoFRhhp6lRBilRCiTghxuxBitBDiSyFErRDiJSGExzi+RAjxlhCiQghRZSwPidOO0IhJCHGjoYqbfz4hxExjW5EQ4nEhxDYhxBYhxB1CCKexzSmE+IsQYqcQYi1wQqIfxtCQbhFCLDXa96QQItvYdpgQYrMQ4iYhRDnwZOTISggxVAjxX+P8KoUQ/7Rtu0QIscz43PeFEMMTtONkIcQSIUS18ZvvZqy/SQjxSsS+fxdC/COJ32KGEOILIcQDQohK4LZEv4VxzMtCiHLjXvifEGJ327bjjd+pzviu6w3N8l1gkO1aDYrxuWVCiDeM++MbYHSMc9pkbF8ghDjYWD9ACNEohCiz7TvF+L3drZ2Psf8c43f50mjfm0Z7njW+b54QYoRtfymEGGMszxRCPCiEeNs476+FEKONbVEmGeO7fmpcv4eB6cZ3Vhvbs4z7c6MQYrsQ4mEhRE6cdie8l4UQFxv3V50QYq0Q4mfG+pjXRAgxTQgx17jHtgkh/imM5zRjkFLqvzb8AeuBo4D/AncY634KzDGWS4Eq4CeACzjXeF9mbJ8DbAR2N7a7AQm8DhQa673Ax8AooAhYClxkHF8GnAHkAgXAy8BrtvbNAX5qLM8APo9xDkOBrcBxxvtZwCNAHtAP+Ab4mbHtCmC5cUwpMNtoryvB77PYtv8Xtt/pMMAP3ANkATnGus3GdifwHfCA0ZZs4CBj2ynAamA343f7LfBlnDaMAxqAHxm/743GsR5gOErwFti+cxuwfxK/xQyj/b8w2pAT47tvA56xvb/EuE5ZwN+ARbZt24CDjeUSYIrtd9rcyn34AvCS0c49gC32aw1cYNwrLuDXQDmQbWx7B7jStu8DwP/F+Z6Z5vWLuMdWo4SOeX+uRD0XLuBp4Enb/hIYY/u8SmCase+zwAvGthFE3Fu0cj8bbX8Dda8VAG8Cd8U5l4T3MkogjAYEcKhxn8S9JsA+wP7GeYwAlgHXJrof0u2v2xuQaX9YAmAPoAboS7gA+AnwTcQxc4EZxvIc4E8R2yVwoO39AuAm2/u/An+L057JQJXtfWsPTI7984H+KIGTY9vnXGC2sfwJcIVt29GRD2mM38e+//HAGmP5MKAFoyOyrTMFwHSgItZno0Zgl9reO4wHdHiMfX8HvBSx7xbgMOP958CFxvKPbO1r7beYAWxs5f6I+8ADxcZvV2S83wj8DCiM2C/0m8T5HCfgAybY1t0Zea0jjqkC9jKWzwa+sH1WOTAtznEziS0Abo24P9+1vT+JcEEXKQD+HXF/LDeWR0TeWyS4n1EddQMw2rZuOrAuzrm09V5+DbgmmWti7HMtMCvZ+yEd/rQJqJ1IKRcDbwE3R2waBGyIWLcBGGx7vynGR263LTfFeJ8PIITIFUI8IoTYIISoBf4HFJtmiiR4HFghpbzHeD8cNUreZqiy1agRcD/b+djbG3lusYjc327CqJBSNsc5biiwQUrpj7FtOPB3Wxt3oTqAwTH2DbsGUsqg0SZz3+dQHTvAecZ78zsS/RaR55YQw+Rwt1DeILUo4QjQx3g9A9UBbhBCfCqEmJ7kR/dFjTrjXhfDnLTMMD1Vo0bq5ve+DkwUQoxECcAaKeU3yZ6XQVL3axzKbcuNreybiL4oTXiB7Xq9Z6yPRcJ7WQhxnBDiKyHELuOzjsf6zaIQQowTygRbblzfOxPtn45oAdAx/gBcRngntBXVkdgZhhqBmsgOfOevgfHAflLKQuAQY71o7UAhxM0o88ilttWbUKPePlLKYuOvUEpp2qq3oTpmk2FJtDFy/62294nOfRMwTMSeXN2EMsUU2/5ypJRfxtg37BoIIYTRJvMavAwcJtTcyWlYAqC136K19kdyHsp0dRSqAx5hNglASjlPSnkKSsC8hjLpJPMdFShTVMzrYtj7bwTOAkqklMUobdX83mbjuy5Aaaz/acM5pZIG4zXXtm6AbTnyd9mJEja7265XkZQynkCJey8LIbKAV4G/AP2N3+wdrOcq1jX5F8qkNNZ4Fn9DEs9hOqEFQAeQUq4GXgR+aVv9DjBOCHGeEMIlhDgbmIjSFjqDAtRNXy2EKEUJoVYRQhxntPM0KWWT7Ry2AR8AfxVCFAohHEJNQB9q7PIS8EshxBAhRAnRGk8sfm7sXwrcivqNkuEb1EN6txAiTwiRLYQ40Nj2MHCLMCZRhZqs/XGcz3kJOEEIcaQxsflrVMf+pXHOFSjTwpMoc8GyJH+LtlJgfG8lqlMLuQQKITxCxWkUSSl9QC0QNDZvB8qEEEWxPlRKGUDNQd1maIQTgYsivtePYU4TQvweNb9k52mUSeVk0kQAGNdlC3CBoT1dQvjk9nZgiDnRamh2jwEPCCH6AQghBgshjonzFYnuZQ9qnqYC8BvPi92FOtY1KUBdt3ohxATgynadeDeiBUDH+RNqIg4AKWUlcCKq06lEjcROlFLu7KTv+xvKjr8T+Aql8ibD2SjVeJnNk+FhY9uFqAdgKcpW/Aow0Nj2GPA+anJ2IarjaY3nUB3pWmANcEcyDTQ6tpOAMSj7+Gaj3UgpZ6Emj18w1O3FwHFxPmcFanT7f6jf6STgJCllS0Qbj8Ia/Zsk+i3aytMoM8MW4/O+itj+E2C9cT5XAOcb7V8OPA+sNUwbUV5AwNUo00k5yq7+pG3b+6j7YqXx/c1EmK6klF+gBM5CKWUyZr2u4jLgBtSzszuG0Db4BFgClAshzOfpJtSE9FfG7/gRSkOORdx7WUpZhxogvYS67uehJpfN7bGuyfXGfnXGZyc70EkbhDFRodF0CkJFjP5USvlRd7dFkxghxCfAc1LKfyfY5zHUfMl2qQKbNEkiVIT7YJRDwiXd3Z5YaAGg6VS0AMgMhBD7Ah8CQ43Rr6YXok1AGk0vQwjxFMpUcq3u/Hs3WgPQaDSaXorWADQajaaXklHJrPr06SNHjBjR3c3QaDSajGLBggU7pZRRAXIZJQBGjBjB/Pnzu7sZGo1Gk1EIIWK6+moTkEaj0fRStADQaDSaXooWABqNRtNL0QJAo9FoeilaAGg0Gk0vRQsAjUaj6aVoAaDRaDS9FC0ANBpN2lHd2MIrCzajU9WklowKBNNoNL2Da19cxJwVFUwaXMT4AQXd3Zwei9YANBpN2rFpVyMAy8tru7klPRstAHooP31qPu8tLm99R40mDSnMcQNQUeft5pb0bLQA6IH4A0E+WradK55Z0N1N0WjahcuhaqvvrG9pZU9NR9ACoAdS0+QDwOPSl1eT/uysjx7lN/uCgJoM1qQO3UP0FAJ+eO83sHkBf/1wJQBFhhqt0aQr326sYuodH/Hmd1vD1jf5AgA0tgS6o1m9Bi0AegrLXoevHoQv/sZzX28EYEzf/G5ulEZj4K2DnaujVm+oVJO97y0uJxiUPPv1Bhq8ftbtbAAsQaBJDVoA9BQWPKVeHU4GFmUDaPc5Tfrw38vhn/tAS2PY6sIc5Ym+s97Lu4vLuXXWYnb/w/sEgsr/v1kLgJTSbQJACDFUCDFbCLFUCLFECHFNd7WlR1CzWb0umcVJja8B6CAaTfqw4h31uvmbsNVew9bf7A+yraYp6rAmbQJKKd2pAfiBX0spJwL7Az8XQkzsxvZkLs21sGtN6O1vnE9zjGMeBS07urFRGo3B6o+t5S2GZ1pLAwR8NPtVB+/1BULOCyaHjOur5wBSTLdFAksptwHbjOU6IcQyYDCwtLvalLF89a+oVY94HqBm5bPAqq5vj0ZjZ/nb1nJDJQSD8PfJ0LCDA0umANfT4g9GuXwWZrtCAWGa1JAWcwBCiBHA3sDXMbZdLoSYL4SYX1FR0eVtywg2fc323LG87D8kbHWRT2sAmjSgaReUjYWioWq5qQoa1L3Zr2ohBzp+wOsPUmlzBz1yQj9Kcj1UaTfQlNLtAkAIkQ+8ClwrpYyK+5ZSPiqlnCqlnNq3b1RRew1Acw21zlJ+4/8p17Zc1d2t0WjCadgJuWWQXQzfPQ8/vBS2+VnPXbhr1vLB0u2hdcW5HsryPVQ3+vAFgl3c4N5DtwoAIYQb1fk/K6X8b3e2JaMJtLDLK/HhYl5wfHe3RqOx8LfA1kVQOgrKRql1790MQODCN1mKWjcn69cAZNHCRLGeoJT0yc8CYgeKaTqH7vQCEsDjwDIp5f3d1Y4eQcBHvU9dyi30oUIWWduCehJN043UbYOWOhg+HU78W9impXW5zPQdGXpfSAMPjVvAO1m/4ZTidUzMbyQbL9Pv+oTXvt3SxQ3vHXSnBnAg8BPgCCHEIuPv+G5sT/pTsxnWfx61WgZaaPCbl1JwmPd+XgscoN4213Rd+zSaSOqMhIQFgyC3FE58ILSpUpTwWuAgKguV8984sYl9itWk76HuJUx5ZX9mef4AwDNfbejadvcSuk0ASCk/l1IKKeWeUsrJxt873dWeTCD4f1Nh5gnhK33NiF1rCAJ3nT4JgAZymMueartXp9PVdCM7VVoSCgep16xC9erK5pG5O2jBTf2pMwEY79iMx6OCGMUX/wBgN8dGQOqI4BTR7ZPAmuT478LNOPzRgTJ8eg8Axzm+YVhpbmi1z2mkgfDWdUXzNJrYvHG1ei0dqV6dRn6qST9m7rpdAOT2GU7AU8gpA6vJkcY9HrDs/pPEOpZsrdUuoSlAC4AM4am5NhXYb3ON2/YdAPXksN/IUk7fe7DaxW0IA299VzVRownHHonuzlGvY34Eh9+K7+g7Q5sKctw4+09kWm45osV2vzpcNEkPpzs/A+A3s37oilb3KrQAyBCy/TZTjq/BWt76LQAutweX08E9Z+7JN785Eq+pAZTrh0bTTTQYcTvH3m2t8+TCoTeyvlbl+z918iCy3U7otxvsWAot9VA4BKZdDjPeptzRn/6iCgCHEF19Bj0eLQAyAH8gyNadVdYKW0KtFqkeCrdUYfRup4N+hdlscSpNgHdv0BPBmu6h0khPUjo6atPK7Wqk/9ODDdfQ/rtDczXsWAbFQ+H4+2DY/hSW9qMYta+ub9H56F80A9jV0ILf5w+9ly2WBlDZpNRsRzA8j0qjsKWCNhPFaTRdQVOVcj+uWK7el8USAHUIAaPNlOUD91KvVevAY927ZWX9mOZcDkiWl9cSCEqd5LAT0QIgA/D6g7iEJQACDTtDy34jndOX+UeHHROmLWsBoIkk4IdVH3b+5zbugntGwJ9K4a1rQTiheFjYLqt31PP+knKGleaS43Gqlfn9rB2ybIOX3FJcBLnU+Q6bdjUx+jfvcPajX3V+u3spWgDY8AeCaRl16PUHcGGFwwcr14eWXULygdyP3S8NTwgnhOD2YY+rNzaBodEAMOcuePZMWPdZ535upNfZkKmW54/BUfd/yvLyOiYPLbZW5pRYy1m2OhaTzwPgd+5nQ6u+MbyHNB1HCwAb/5y9mql3fMSOuububkoYXn8QF5YfdLDeSvLmwkdJn/70L84LO8YhYJfLGFVVrgJ/+gk2TTfy2V/Ua6CTk61F3meD9gZgc1UjLf7wnD6TBtsi1s34AID8Adby4KmhxSucb3RaMzUKLQBsfLZKjZSXbUsf3/lPV1bw3aaacAHQYsUDuKQfh8sTdZxDCBoxXO8+fwBevCDlbdVkII5OzghvxqrsdwVMPh8Ou4XVO+o46J7Z/GbWD2EFXg4dZ0vuaLdZjrWZM93Z8IuFABzh/LZz26rpvnoA6Yg5ueRPk+yDgaDkoidUBaU9hPXgSJ/yAtrV0IIHP8KVFfP4ILaHatUHqWuoJnPpbM3Qp7TnBZ6pyN2OZLzI46j71b33yoLNvL9YpYb41Y/GMbZ/RMnSXy6Cxa/C4H3C15eNZnZgL8qEjmrvbLQGYKPF6Pj9wfTwMvjjm0tCy+4YGsC3G6vw4KNvUXTtX4cQKg7nlAdT3k5NhmF3CvB3srnT0ADu+WgDZz48l5fmhzsg1HmVM0NZfrTWSulIOOR6cER3SwX9hjNUVADp8Wz2FLQAsGHWJw2kiQB42hb967QJAGnEAdQ3+/CIAHm5OVHHCmFoNHtfAKMOgz46TbTG4HtbPv4UaQDNqA7+9rdiF/gTtC2oa/J+R1Ai6hkqdJGjzkQLABumBpCOBSjcNhOQ39AAGhqVIPBkZUft7xCCoOkvPe442LkiFDWs6eXY3TLbqwFsWRg7ytyo9GUKgHjsNbQo4fZIXGXDARhAFQ4dENxpaAFgY2edGg2lgwYQ6Y7qtLmB1tUpW6ivUUX4evKiHyaHsCnLk88FVzb88EpK2qrJMOw1ItorAB47HB4+KHydlPDZ/dBnPOvkwISH7z6obQKAAvV5Q8UOxkXOHWjajRYANhoMD4V0mAP489vLwt67ULbTGplLXr0yDfkbqwFw5xZHf4AQhE4ju0il463blqLWajKKoBVU2CkmoOZa+NeBMP9xqFpHcN/L8EX4l+w3srRj31E6mo3Bvtzmfor93Gs79lmaEFoAGNhH/bsaWqhq6N5i1M0R+c/NQLAFwXGUNqyFlkYChgYgsuNoAPaQ+ZxSFaWp0UjbvRWRQqRdbJkP2xfD26qsoy9vQNQuMy+exmc3Hs4xu/fnqUumtf07XB4u8d2AmwA/anwrtLqy3ktNUyecQy9FCwADu93/7neXs/ftKQiTbwM5bmfYe3MS+HsxDgcBKP8e2Wy4xWUXRh6OIDwbL7ml0KQFgIZwDSDQwc7T3wILZoatanAVh73PcTvJ8TgZWprLIz+ZGu7/3wYeuvY8tsoyXNJq8z53fMRBd3/Srs/TaAEQIh3MPna8hkC690xV2ct0A13imKB22LqI7Caz3F60vTVsEhigbAxUrAyvJaDpndjnADoqADZ9BUtfD1t11ZvloeXz9hvGN7ceGXlUuxjXv4AstzNiZGO5lmrajg4EM/D508vzp8UfZLeBhZw1dSiHje+Ld2EFzIYqd198fg+umk0Ed1UQROAoGhJ1fCgOwKTvBOWj3bADYuyv6UVI273enlQQdgGy6ZvQYn3OYOb5R/NVpVWZbvLQYgqyw3MBdQyBMNwbgmk2aMtEtAAw8AXTSwB4/UGyjPzn/QqyoVA9RE6Xh2r6Eli/jEJvC+WOEgbFigQWhGsA5jxBc40WAL0duwko2I7Rs11rMATA0d57OHzaoTzyP2uCdtZVB4QnfOsEpO1/VaPWZjuKNgEZ+ALpNZrw+gLhBTCMB9Xl8rAmazf6bZvNgY7FbJL9Yh4f5gYK4QJA07sxR/AOVzs1AJsA2PgV1bkjWSmHsqaiIWy3vYeVIDq7ipeKcARgZ70WAB1FCwCDtDMBBSwNAAg9dC63i/fzTsZBkAGiivx+I2Mer0xAdg3AmChu1vlUej2mF5Arp31zAPZjvDXU5CiN8qNl2zuhcYmRyr0BCI+V0UVi2ocWAAb+FJmApIxfwWj28h28vmhLzG1eX5Asl80TyBi1Od1ZrBIjCAplvRsycreYxwsBYSbS7GL1qjUAjWn2cWe3zw00wmy0cPD5Ubt02O8/LpZGYXf/TDcNPlPQcwAGLf7U3ED73/Ux2W4nn95wOKAEwty1lew/soyLZ84D4JTJg6OO8/oD4RqAMepyuz1Uel1UlEym/675BI1865FEawDaBKQxMAc7HdUAjv8LTPox6z/fAawCYHBxDm//8iAKO3XiNwLjvvb6rcnoZn9A1wxuB/oXM0iVBrC91suGSquI+4dLt3PeY1/zn682xD1GSkldsz/CBGTOAbjw+gLMG/0LHvWfgBwd38UuTAMwC26sfFeVA9T0XswRvMvTPgFgag3uXMgpxhcI4nQIXr5iOh/+6hCKcz04Upawx/ICMpM3QnTgpCY5tAAw6KoEcNuNfEN/eGNJ3H3WVzayo87L5GHF1krjoXV7smj2BdiSP4k7/eeTlRW7FoDTIcKFmlk0Zs0n8OfoSE1NL0IGVK1ep6d9k8DmAMIo9egPSjxOB/uOKCXXk1qjgrTJFa9t3s4uDDTJowWAwUfLVBbDUX1VacXISNzOwhVjZPTpygo27Wqk3gho2VCpvCl2G2iL8DUEgMflptkXDJXXi6f2Di3JZcPOxtjzD0EfLP5vR05Dk8kEA8oDyJ0DLQ2t7x9Ji1Exz6OelUf/t5amLhuBi5gmoK77/p6FFgAG/5qzBoAXLt+fQ8f1Jcvd/p+mxR9kxM1vhz7TTm2MvCUXPfENB987m9Mf+gKAbTUqQ+OgYlue/6AfhIMsj5vy2mYafQGEiC1QAIaV5lLn9YdHSeaWWcuf3N6OM9P0CIJ+cDhVfqimqrYfX2skFYwRgZ5q7F5A9lH/B0vK4xyhSYQWAAYluUqd7VeQzcg+eR1KCW2O5O97f3lonTkSr2qMb3Ndub0eUMnoAMrybDnVAz5wuBhepqIs/zVnDVIS188626M0mDDV+JL34dh71PKuteCtb8NZaXoMMqg0gPbmh6qzBEB9l6dhsARAs00DiCw4r0kOLQAM+uRncfwkZRt3OkSHwsxNddT+EeaDUt1K9OLW6ia+WltJtttBtt0MFfSDw83hE2IHfkViTiCHTY71GQv7X6G8NyBUvEPTyzC0SXJKoKkdXmF129TxeX1DNTS6DCEQ0tIAcj1OCrJc1DZrx4b2oAWAQZMvEOpwXQ7R7uRwW6qbqGuOHuVf/ZyqxlXV2ML4/gWcvne06yfAAXd/wmerdlKaG1FRKegHh4v+hdHVv2JhnovdThoip0S96sRwvZNgQJmAXNngrYF3bmjb8XXbIL8/OF3UGR3vYxdOTUFDo7GngvAFgridDgqyrXZo2oaOAzBo9gVDnabDEZFJM0lqGn0cePcn7D4oOj3zpysrAGUCKs51E2jl84sjBUDABw5n2OT0GVPi5/TJDmkAMVRjlyFEAl08etOkB9KYBHYa99g3j0KfcTDtsuSOr90GBUpbNgc7Bdld1ZVYJiBfUOJ2Cgqy3TEHXZrW6VYNQAjxhBBihxBicXe2A5SpxOxcnUK0aw6gukmNqJdsjZ9uobqxhZJcT6saRmlehADwe8Gdi9M26XvzcRPiHp9lnEtM/2gzeZzWAHonQb9yA3XZ7rGNXyV37OqPYc3HIU+cxz9fB3StABDGoxMISJwOQY7HyfebdYBje+huE9BM4NhubgNgmoDUz6E0gOj8InNW7OCZBAFciVzRzA69qtFHSZ6bQCuh68W5EZGUvkbltmejb0HsGACwNABvrMkxc+SnNYDeSTAYrgG0hc8fUK8DVZ2Kj5ereaSURv7akMLSAPxBicvhYNGmasprm/ly9c4uaUNPolsFgJTyf0C3l6kKBiWBoMTjtDQAiIikBWY8OY/fvhZfWalPYIecMqwEKSXVjS3R5p0YRAuApigBkIgcwwuosSWRBqAFQK8k6AeHA5y2AUSyDgENFTBwMhxzF6BSP7gcgqGluYmP6zSsSOBAMIjLaWnEa3a2I6ahl9PdGkCrCCEuF0LMF0LMr6ioSMl3mOYY82aSxg3W0NK2iaVEE1ENXj8NLQF8AUlJrpvfnrgbp04eFHf//KxYGkDyD1lRjjo+Zr1Uc+S3djbMfyLpz9T0EEKRwLZ7LJkssTtXQcVyGHUYjSIbXyBIky/AOdOGpq6tUYTPAdhNoqlKPtGTSXsBIKV8VEo5VUo5tW/f9tUSBVX0/au1lTG3mSkTzKCquWvUfk9/ub5N3xGvNN3g4hwaW/yhQvPFuR6GlORy9RFj4n5WriciEtnfrLI3ApccOJIbjhmfsC2mlnH9y9+xuaoxfKOpAXz5f/DWdVEl9jQ9HDMS2F5IqLUkgZVr4J+Gp8/Yo5n4+/e5+Ml5NHj95KU4/YMdafX/BAISt8Pqwn7/erdPJWYcaS8AOovb3ljCOY9+xfLy6JGOqQGYo4nbT90DgB1t8HFevaOeVxZsjrltSEkO9V5/KMK3xOic++RbD+CkwUVhx9i3AWEawO9PmsjPD48vPAAKsqyH8sHZERHJzojP3jwv4WdpehhmJLD9PmiuTnzMhi+s5REHAvD56p14/UHysrpOAAjC5wDsGoCuENl2eo0A+GCpChWPlTTKnJA1NYBx/QuYNrKU1xdtpdkX4MZXvuP7zdWh/f0xEscd//fP+N9Ky0R152mT8DjVzzu4JIdtNc18bWggew1Rnb1ppoHonD5n7hPh4umtb5MJyJ6N8flvNrJgg22qJTciV3uyHiCanoEMGiYgW8fdVAXbvot/TO1W9XrBf6NiS7pSANjjACLnACD2s6mJT3e7gT4PzAXGCyE2CyEuTdV3maHisdI+m/WAXU7r51hbUU9Nk49LZs7jpfmbOfmf1gioOYZnTUvEjTekJIc3f3EQfzpld6oaWmhsCfDQnDU4HYJ+RjCXEII7T5vEW784iHvO2DN07Gl7D45O8uattXL6t4MX522y3piBYCYf/q7dn6vJQEwNwJOv3vffAxzuxAkCG3dBVhGMOTJqrivKXJlKhKC5JUCzLxDSAKbYsuZWx5rz0sSlu72AzpVSDpRSuqWUQ6SUj6fqu0wBEKvwi+nzb0+sZtYbXV5eF7X//PW72FbTFPe7rj1qLPuPKmP8gAIunD4Cp2GnbPIFQlqByXn7DWOPwUWM6ZcfKqAdleBNSjVJlx0dYJYI+4MRVl1MCBiXFt63mu7AjAQeup96f8AvlFvn+s/jH9O0C3LVwCEyoeHYfvmpamkUXr9EIPnjm0uobfbjcgieu2z/0HZznk2THEkLACFEV/l5pQRzhB5LA/AHwucA7GTFSLc848l5TL/rk7B1+TY1+NqjxoWN4O893kr7kKhqkRmIVpof4Sbqa1IpnLPaJgAev2jf+N973otwyQcwObqcn6aHY0YCZxfCbTWw1zmw20mwZT5Ub4x9zIYvoWQEEO7tdsmBI5k6IlXlH6PxBVQ+0LlrKvluUzXz1leR7XYy82J1r+ucQG2jVQEghDhACLEUWG6830sI8VDKW9bJmDVDfYEg9763PFSRq7LeG7Jpum2jc1MWmBO3rWG354exeT6lD07gH4XPAokFgOmyObpPxIiq3ii2bU/nnAQltmjimN87bD8oHaWWdUxA7yFouIHaGXWYep1zT/T+fi/UboHhBwHhAmDCwIIUNTI2QQlCSCojRvrms6vnANpGMhrAA8AxQCWAlPI74JBUNiqVVDf6eGjOGn732mKaWgLsc8dHoepcdg0gLBd/EuTEs4NuWQDAyS1vU0ptlAnIjqmljOiTF77hrevUa9/4qR9aIy9e+0ytwhtt6tL0UEwTkJ3S0ep10TPR+zcYEbb5yg3bzLtz+pTBnJIgliUVBCQIZNQ8hGk2bW8Sx95KUiYgKeWmiFUZW36nwubauaZC5cP/YrXyzrHb3p+5dL+kP7OmycfqHXFy65veE8Bujg0JNYArD1UPYZhN1e9VLnj994Ahbc+4ePh49dD64qWeMOcVNs5t82drMhRzEthOdiGMONjYHjGKNvP/56lU5Fuq1fzXr48eHz631AXES/tvOnB0VWnXnkIyAmCTEOIAQAoh3EKI64FlKW5XyrjrXatIy7mPhrs/2r2AzMIrduJ5O9z7nvWZ/zxvb2tDSwN88bfQ20IaCdVvCQagLryK0Rn7DGH93SeEmW7Y9r2q23roTdEPbRI8efE0PC5H7JxAACOUWk9V/BxHmh6GjGECAhh1qLXdjuke2n93AOavr6I4103/BLmoUkXQ0ABMLpo+HAC34Q7akUJOvZFkBMAVwM+BwcAWYLLxPuOJjNy1awCxKm3Fs/Obo46iHDcn7mlTief9O2y/QtHI2ooG8DXD3cPhr+Nbr8laYQiXAZMS75eALJcjdl0AgIJBqrhHe0oDajITMxI4EnNdMGIitdGIISlU93ad18fovvlhA6auQtpyAQFcb0TEm+bbuJquJiatRnBIKXcCGe8qIkR4xgOnIzrlc6QXUGmeJ1SeEVQHH2tS2DTrRMmMiMCtYgwz0auXWoW163dA6cj4Da83tIQO1F/Ncjlj1wUAlRQsu6j1SFBNzyHWHABYAiDgC0886GtQcQJG7qD6Zn9SCQ1TgSQ854+ZhiI0CRzDy08Tn2S8gJ4SQhTb3pcIITIug5g9ZwhYNYDtRPrf337KHmHvx7TV3zm7OLQo8/pzdd9FfHZBMSx/y9rnP6cqIRCPunLVQbuTqwQWsxnuBBoAqOpO8/4Nu9a1+zs0GYQMKK0vEofxTERqAL6msMFMnddPfpfl/w+nLD87pAFkux2hiPfQJLDWANpEMjrcnlLKavONlLIK2Dv+7ulHMCijInWbYqRJznKH/xwtgfB9cj1OCm03vlkvYEiJejj+ePLu4R9o5ts/8wnEvpdQUL2Mod5Vat3Ayeq1aj2s+1/shu9Ypjrm9uRtt5GVaA4A4IBfqtdXLunQ92gyhIAvPBOoiakVBCOeDV8jeCwBUN/sJ78LE8DZ2W1gIUOK1WDIXjPbrSeB20UyAsAhhAjlDhBClJJhpSQf/t+aqHUNMQSAWQ/A5IDRfcK0gmy3E4/N66GiXnXwZv2AIyILtvuMaOERhygvHoDP/qpef/wknPg3tRzP/r76I/VqlN9rL1kuZ8wcSCEmn6e0gMrVHfoeTYYQ8EYnBATbHEBEOoXmmlAZ0W01Teyo8zK6X4SrclchRGh+Ltv2LLr0JHC7SEYA/BWYK4S4XQhxB/AlcG9qm9W5LNyQ3ARnpAbQvzCb1Xcez/j+KtilJNcTJhDe/l65x5m5hNyRk2JmcJUrCwbvo5bNSMucUtXxQnz7u+mbf/4rSbU/HlmtmYCEUOkAvLWWz7em5xLwhZeDNLFPAksJ9+8OC5+Gpa9DlTIPrq1QTguTBhd3UWOjcdhMQCYuw8Tr0wKgTbQqAKSUTwOnA9uBcuB0KeV/Ut2wziSh+cNGrLQPYAW+9MkPr+X7xzeXApbdMWwOob4CPrhVLbuyoXAg5BsjeYdb2fVdWcq22lQd/aVSKve73D4d1gCyW9MAAEqMiejaLR36Lk0G4PfGNis6bXMADRVQuxne+pWxTWkMZgzAoOL2z0l1DBFytrCbgKw5AG0CagtxBYAQotB4LUV1/M8Zf+XGuoyhJWkBENvPfqvh+dOvMBt7H7+XkbzNFwgiRIQX0fcvWsvmg7X7qer1V8ssl6Hs4tgawNcPw8r3YNj+0dvaSKsaAECWMcHtjRPQpuk5BFpaMQEF4L+Xq2Uz/chhNwOwaFM1+VmuNkfKdxpChDqtLHe0CUhPAreNRBrAc8brAmC+7c98nzFETgDHI54GYDJlWEkoFqBvQRbBoJlfSFUmCosd2LlSvU481ersf/QnuOb7UEg9ADnF0RqAvwXeUw8cx96dVNsTkeN2xq4NbMdj5HRp0QIg7XnyePi/tkeFhwi0xDEBGR1qwKfKhYKVK8qTR73Xz3NfbyQvyxlt7uwyBA4hjSULsz2tDnQ0YcS9ilLKE4Xq0Q6VUo6y/Y2UUo7qwjZ2mP4FyamridI0gHId/etZe3HgmDKmjSiluknFCPgD0YUp2PotjDocznrKWufKgpLh4ftlFyu30Af3h9euUkFi239Q20YfAcUdr7dakuehqrGVNLkhDUDnBEp7NnwBlavad6yUhgYQSwDYTEBmx2+6hLpz+GadSpmyvbYbEwcKYcv7Yw3sst1O+uRnsb6yMd6Rmhgk7PGk8nN8u4vakjISlU8c19/y7W9NA3A5Hew5pJhnf7o//QqzqG5QcwO+QFCNQPwt8NiR8PWjsGMpDErCWzanWL1WLINFz8IDu8NjR6h1J9zf+vFJUJbnYWd9S2JTmEcLgIzjgT3gkzvadkzAGAjEFAC2SWCfEfDYaNTRduca5Rjp8gRw4YiQGdYXUdtjRFludP1rTUKS0eMWCiH2bX239MWR4CyPnjiAJy/el8sOHtmm0PbiHA91Xj8bKhuo8/pVLv/aLSqn+rs3qIdo8JTWP+iEv4YLikabF06iCOE2UGDELny0bHv8nfL7qclq03SlSX9qNsH/7rPcjZPBFACuRHMAfuURBrDLcKF254TSlf/yyLHtbHAnIARZTsFeQ4tDtbtNPC4HX63dpWMB2kAyPd5+KDfQNUKI74UQPwghvk91wzqTWIVeTFxOweHj+3HrCRPb9JnFRiTxoffN4b8LtzCsLNfKmmiSjAZQOAgunwMXvwtTLmpTG5Ll+EkqjUR1Y4JyeU63Sva1I2Pz/PUezCBCk/oEgj0SU1jEmgQ2awTvWhs9F5RdHEqLErf2RZeg5gBe//mBTBsZ7ovy5RqlrcxaqD3ZkiUZAXAMMBo4AjgJONF4zRgKsuPfsFHlF2Pw1S1H8vGvDw1bVxyRSmJkWV60ACgcTNIMP8CyuwJckKA+axsx87b8ZtYPiXcsGqpGlZlMzRaYdUVs19qegtOjCrgcc6d6X9cGAWAmFyyLMY1nBHvx2pXq9ce2+avcMhZvqaFPfhZled2TBwgwknol3kW2toMmRDJxABuAMuAU4GSgzFiXMQwuzuG5y/bjtpOiR/n2Uo7xGFCUzei+4XmAIkdBI/vmhadSKB0dIztcK4w5UrndnfBXtdxJ2IvBxEqBEaJoCNRsDs+al2ksmQXfPQ8f/aG7W5JCpMrlM/wA9baxDcF7Zgry4uHR20yzkAwq7XXiKda23DK21zUzvCw3ZqbcrkMQTwL0yVftt6d81yQmmWRwvweeQgmBPsCTQojfprphnc0Bo/uENAH7qL80v305zUsisiGG+UX/Zitc+UXbP3TAJLhxLez703a1KR72B/bimd/EnygrGgr+ZmUCmH2nNRGYURidw4KZUJ3h2kw8ZBAQ7avmZkanx5oEdtnu4QknqAHMxe/CHmdCbhl1zf7QfFK3EZnW18YLl6siTglNnZowkjEBnQ/sK6X8g5TyD8D+wE9S26zUYE4G26N5S9uZ1jbSBDS6wPjM4QeBJy88nW4acPGBIwD4au0u/vD6ktg7mS6nH/0BPr0H5j3WNY3rTOypLN69sfvakUqkoQG0RwAEbOlJIrGvM73Chh8AZz4ODge1Tb6E5tSuI7YA6FfYXdHJmUsyAmArYP9ls1CFYTKOHTH8lycNLrLeeOtg1pXQUNnqZ/W33WwPnL0Xu5v12vf8cUebmRKmDrcmzD5eHif9dNEQ9brdEBBtmVxMFxoqrOUOptBIX6QaCZvlPJtrkj/Un8AN1GV7zD3hyd5a/EG2VjentQZgz1AqM9mM2YUkIwBqgCVCiJlCiCeBxUC1EOIfQoh/pLZ5nUttc7hq6HE6KLKP5JfMgu+eg09ub/Wz7HlIpg4vVSlzAdzdlCWxFVqLcQCUCQiUCQigMQOrhNWVw6ApagTr6pkjwm3VjazY3qBG7K7stgnqkAYQ47ex15yIKGY0f8MuWgJBpo3o7iww8ecfHA7BDUaFMF0ZLDmSEQCzgN8As4E5wK3A66iUEAtS1rIUYOYJOXea6uiO3C0ifbN509sKuSdD34Isq7SjJzfxzt2EXWDF36mYsAcs0qspE6jfrkb+nvwem9aist7LxirDnXPIvrB5XvIH+xPEAYRpAOFOD2YW0EjXy+4hfuduzu/t/acPktYCFmyoYtOu3hlAlkxJyKda2ydT+PkRY3A6BNf9aByXHTwqOqGVaUtNVKe3pSFKPc52O20aQHoKgMhU11LKaG8Oh4OwhysTBUBdueoUs/J7bGI7h70wYsFA2PxN8gcHvKogfKySkAlMQGYQWJ92Ok10GglMQGDlBGpoCVDd6KMkCZfVM/71JQDr7z6hc9qYQXRXRqduoTDbzY3HTsDtdDCqb370qNicQPTFGQ3UboU7B6noS+CdXx7Mv843on1bjGM8mWECSipBXhs1oW5n2/fKJVIIJYjjXceOEFktqxsQSIKmAMgualvMg98be/QP4W7L2UVhmxq8ftxO0Wq+rNQT3w0UwG1rX3ltJnqxdS3dfTXTi22L1GsghhuZvwUePVwtf3IH/GU8EwcVcpwRZYvP0BrSVAOIFHaN3jgd2dhjrOXmapWgLlPcKTd+BUBTyQTVyQVaSYDXVqo2qAHA9y917ue2EYFE2gWAtxaSLYYeLxFcJDklYW8bWwIq3Ul3E6kBSGk5LQAeW1JGM3I5EcFeXkBGCwA7pjdFLNtxxTKoL7fe25fBpgGkpwAojghca/TFEQDDp6tXczJ70bOqNkG68+U/VQ4m4FdLRqlOLpYg7wgr31NxEv+9DBp3de5nt4EwDSCnWMUFJDvf0bjLSkCYiIh9Grx+8pIImkw9ERrAvH/Dvw6A5SpnpT1NdTJ1QOI+B72ERAVh3hRCvBHvrysb2WWYqW9jzQHEsifbO5iQBpCeJqDiiHiHRq8/9o7m6Co3HSb72oBZfQ1YtCOgcht1tgaw7n/W8r0jYcV7nfv5SeIwkh1IKS1TTbyyopHUbLa8vWJxkFEBLGISuLbZR64nDTWAtXPU6wuqvKpdADTbOvclW2v43WuLQxPDW6qbGHHz28yO5xLdS0ikAfwFVQ94HdAEPGb81QPRVdZ7AmaHHikA1s6BmcerZXt4vD1xWpprAJG224Z4KSHM9L99bBkfN8/LqPQQ2+oDhgbQyQJg+xIYcbD1fvWHnfv5SaJMQA5V6jS7WK1MNhbA9JKKx1F/gNtqotKYzFtf1X1VwMKI0AC2LrKWg4GwTKBXPruQl+cr8+WFj3/Df77aQKVhFvpwidLg//3Z2lQ3OK1JVBDmUynlp8CBUsqzpZRvGn/nAQfHOy6jMTUAXwPU2jxgnj3LWt7zHGvZsDmrY9LbCyiSuBqAWYJyjzOtdZu+hm8eTX2jOpNUmID8zVAygjO8f2BpcLjSAIJBePUyeP3nnftdCRCGBtDUErA0gGQngr11VgRxktQ1+9jV0MJBY/q06biUEKkB2IV8xfKoNBC3v2XU7TZs/Q5DsNU1q/t/g+H+mVScTA8kmbPOE0KEUgcKIUYC6Wnn6Ch2D4/7J1jLAVsE8bD94Zi71HKDTX1saVApdmO516UJPz98NKP6qksXVwPY7SS4ZXN0Mrqlryc90fj12kpmfbu5I03tOKkwAfmbwZXNAjmeh/wnq6LpS/4LP7wE3z7Tud+VADUH4OB/qypsJqAkNQBvHWQVtOn7thveNAOK0iGwLkIDEMKqW1z+A8EITdV0Ww0YAsCsIuYw4gVMgTG0NDMGbp1NMgLgOmCOEGKOEOJTVEDYtZ3x5UKIY4UQK4QQq4UQN3fGZ3aIoC/cxGPisE1+5ZTA9KvUsr0YR4z4gHTjhmMm8O8LVS3ZuuYEo+OsAsjtAyUjrbzxG76w6hS3wtmPfsV1L37X0ea2i5t9RiK9VJiAfM2haNl3girxGK9e2rnfkQSmBnDNC4usydpk5gD8XjWYaaMA2FZjCIB0yLUTqQF462GSkX5l1s+4sHABlx1sFVIKGPuGBEBA8vTc9Xy1Njzdy+od9fh7YSGZZNJBvweMBa4BfgmMl1K+39EvFkI4gQeB44CJwLlCiLZVZelsgn7V2e97WbgbnJnXf/+fW7ZRM51ulZEZu7ES8tJARW6FwSU5uByC+RuqEqeGdrrgmkUqs2m/3dW6bx7pkja2C+FkzYQreCGgymkGO9sEJCX4m3j4SxUbEYz16CSTPXXNbKhY0aGmOAwNAGibBmDu00YBUG4IgIFFaTAHICVUb4C5DymN1Neg5kEG7AmAZ+Pn3Hispb37AxJ/IEiTMSHsD0h+//oSPlsVnUL7re8zMPCxgySTDjoXuAG4Wkr5HTBMCHFiJ3z3NGC1lHKtlLIFeAFVc6D7MAVAbqmyqZomoeYamHY5HHunte8Jf1WvZvRww041ak5zslxOBhXn8NzXGzn5n5+3foDTpcwpcWjxB3lw9urEGkU7qWpo4cOlSeS5CQZABmgMWrdzUHSyCci4zrU+y8TXcEaE2ae6lTIZjbvgP6fCg9M61BRhL3mSZQiA938D1RvDvysyfmPrt+q1X9vGWaYA6FfYzVHAYOU9ev8W2LpQLWflw3kvqt+iYWe4K2ggyK2zFofe+2OYMScMUALx2hcXpazZ6UoyJqAngRbAcBBnC9DGStQxGQzY79DNxrowhBCXCyHmCyHmV1RURG5OnmAA7hoGDx8Uf5+AHxxuY/QvVccfDKpX09vCxBxFtRgCoHFnxrhOmqmsV+2o58V5G1vZGxhtBMC5okeAz3y1gfveX8HTczu/RtBlT8/nsqfnh9IQxMXo6Gu8ludK0OG28t50Bh+oEhh7OSyvkaaRR8MpD1r7vHND/GM/uQNm/9laV9/+e9khbKkg7AWvZ99lLf/fPvC38Jq5rPlEvQ7cs03fV17bTEmuO7l8UqnGLtTnPa5e+05QpVUHTIKm8ASGFXVeXpxvdTP+GIFffQsswdbbsogmIwBGSynvBXwAUspGEqXk62SklI9KKadKKaf27du3/R/UXAPeGij/If7DF/SrSVzT/NNUpY5BRkVGhgSAPX9QG70ruout1Zap4qZXWykTCXDwryF/QLhrqMGGSuUyG6tz6OjDNH+DepgbW+J4LJkYnUKVTQAEHB7wt6FYemusnQ3AOjkwtKrFH7RSaEO4S6J57r5m+PL/1HzRvH9b25vaH0jmFBCUtkfwQiMsx54K2/z8Fls6jEXPqVfDbBQMSv7y/grWVCQOItte6w1Lf96t+G1mtpVGHEZfw+STld9qbYRYwWGj+lhzd5VJRA/3JJIRAC1CiByMqXchxGggOrF+29kC2CNShpDKOgN2D594dW9NE1COMZKvWG7VBogSAEZnb7rftdSnbQxAJDvr23j5sgpgyFTLTdaGmVDunneXhybaTLxJRGLG44vVlo22IZ7Lqokx0t9l6xsCrlwlGAKtHJsMtVuheiOfOfblAf8ZodW+QDC87rO3Rs07BHzwx2L4/AGoWh/+WeaEZVty+EfgFIRSQQSDEkYdCuNPiJ27yTSZ7FytUkbsflpo029fX8w/Z6/mV62YPmqbfFEFkLoNv+3ebdqlCuMUGELZk9dqRLTXb/UDZ08dykPnT+GW43cLrVuwIQNToHeAZATAbcB7wFAhxLPAx0BnlFqaB4wVQowUQniAc4DURRhLW2cUoyNT632GADA6+xfOg+9fUMv9I+ymhYOUScQsst3SmPZeQCa/PzH8XH7+3MLWD3K6Y06qmqP8lkCQRZuqw7a1RwCsrahnTUV9mJdGfby8RSYhDcBa5XcaI1YzQrsjGDEhsxxH4cWKqP5hS40lAOxa4wajHOhHt8FDhrfQOc/Bb7bBflcY+1W3uzkOWyqI+z9cqVZ68qxYFLud2xQ0bxsRvqMOQ0rJvz9by3NfK/NfliuxaafO6yc/Kw0FAEDpKDVPBUYKcHW9bzp2AiftNSjq8KYW67cZ0SeP4ycNJNvt5JUrlIV7R11njG0zh2S8gD4ATgdmAM8DU6WUczr6xVJKP3A18D6wDHhJShmnVmEnIG2dSDzvkKCRQiDbZsoxMn+aXgYhHE4oGaG0iYBfuddFhM+nK5ERnW9/v40ZT37DS/MSJH1zuJWAjOBz20h9S3W4ycXbxjwrUkqO+Oun/PjhuaFAHYD65uRMQI0BqyO79xMjDsFuAmkvxih6RzA8Q+YXqyv5fkcL45tn8skoY0zUuAuWvBb9GUP3Uxqi6bPeoWprVjK4b9YZph53juWSbM+C6q1Vr2Z8yp5n893mGu5424piH90v8X1b7/V1fyUwE3MOwGEIpMFTrW2efPW7NlRy5WGjuev0SVGHN9nuydBU+rrPmPzx+eTT2OZ7NtNJxgvoY2A/KeXbUsq3pJQ7hRCdEhYqpXxHSjlOSjlaSvnn1o/oyJfZRkXxvEPMOYBIW/7AyVGh8YAa9TVVW6PMDNEAzKIZduasqODGV7+Pf5DTHWVO+WptJWsqrBF2c4RbaVIpp22YE3S7GlpCEZsA9a2ZgIzr6ZXOUERnozQm9jojJfTWhSCcrAuGFxBaW1HPyf/8Ai8e/rPEsD/VbVOV5UYcDP1tHZDZ8Zt5eN78Zbub47AJgFCdB3eOmvOYdWUoLw5gaQDeOhh1GLhzWLcz3Ezy/DcbeeeH+C6Qdc1+8tMiERzRGsDAvaxlM4p96WsA5MXIXfT03PWhZbc5gf7Uibg2fclNrhfC8gf1BpIxAY0EbhJC/MG2bmq8ndOWYBIaQMAwAdk1gJvWw08/ir1/bqka8bXTv7q7cBoCwHR/y3YncRs4XFEaQOTI/MZXv2fhRsuG6m9jWT67ycge0dnqHIAhAJqlK1S0vAlDACQq7pMsFSugbDRVwXAB//U6ayJ3A4Yduvx7FZQ15ii40uZmawo0p0ulC5HBDuRXsgSAx2kTAL4mVdJ03afWrs2GBtCwMzS3FZkuAeCqZ2ObAbdUN1Hd6EuTPEBY19M04xZak/LsdpKKzzG8newFj/YZrkx0pv//yD55/GT68LCsrj9xfcSg7XNS1/Y0JBkBUA0cCfQ3MoQWtbJ/etKaBiClMhM53OH5fHJK4vvB5xQrm69ZSCYD4gCAUFbHsf0LmDailGZfEiP1GHMALiP3+n62MoGnP/RlaNnXRg3g+a8tl1R7kFpDa15AxiRwc9BFP8Olz4txzTojFqB+O+T3TyjQKihRUdPbjAhocxDxq2VwTURU9DFGPMmmr9WrlG1yC7XPAZjXAHdu7HNtrlFmsOoNIS8u0632xmPHt/pdWw2z3u6D0sTDbZIxCW8GXdqfOSGgbEzMSnbXHBnuwfajif3JdgpY+LRaMewAAMrqlnd6k9OZZASAkFL6pZRXAa8CnwP9Wjkm/QibBI6hAZgagsOlbqTDfgMXv5v4M3NKlCeCOYrIgEhgUHVdf3fiRO44ZY/kKzw53FGT56bg+F3EpLJJWwpzf75qJ39+x7JLN7T4GWyMOuuSnAPwSkfIp9uHM2xbu/HWwZYFUDQEXzDIYeP7xrGHCzUPtPhV9dacDyocpOaK7Jh5ltZ8Ag/sAW9dC38ZE1bYJDE2DcCcwHVHjNBLjHQI3lqo3aLu/9LRgBIABVkurjpsTNghsUxtXuMap0UMAMBx98ENa+G4e1Tn3zdCiOUUh02wm9bOycOKw3Zr9gXg30fCR4ZhY8bbBHAQ9Hd+QGM6k8zTH6oGIqWciZoM/iBF7UkdrZmATKFgTpYddhMMPyDxZ+aUKr/kt65T7007b5ojhODSg0ZSlOtOXgA4XaozefemUMoD014aL098WzSACx7/Oux9XbOfohw3LodIwgSk7MLNQVcoX41PGp10ewSAlLDwP3D3MHjuHAj6CQ4/CClh76El/HDbMWG+4zEx3BEDQcnML9aF25bzjXTMn96jnAgWzFTvN3xJMjiQDC3NY3TfPKuiVaQA8OSrv+Yaa4Bi3J9NLQFys9Q1+/C6Q0KHvBMjFUJLQLU7bbJlurMhrwz2OANuXAP5EWPR7OKwvEgzL57GkRP6ke9xcaDjBxyoe7LJ67ciifuMA4eDFtzsURVh7q3eBM+fZ5nSehiJCsKYOt/LQohS8w9VH+D6LmldZ9KaCcgc3TraMNlluv7VGKaLDBEAdiIf7MhOOxCUvPPDNqsE4dcPw4p3APjHJ6sAyIkjAGKF3SdLeU0z+dku8rJcSQgAJbybgk6KDH91H6YAaEccwKZv4I2rVee5Qdnx/Xn9Acvk8vBP9gk/RgLnvQRjj6Fp0vm0TFT+/m98t4Xb3lzKQ7NXW/u64pRkrE+2OIlECoHH5bQm2iPSkJdv36bKYn71EDxnxB7kqvu12RcIjejH9rfmrWJpNqYGkJXMPFE6YGoAxvzKIeP68viMfXGs/oBnPXfxU6eqHHbhFOPZ3fsnMEPdz9ucg+nr2xKeQuPD38OKt2FFK9aADCXRVTXCBlkAzDdeF9jeZxatuYGaAiBB3pso+u0W/j4786ZHtkcUzo6sFfzkF+u46tmFfN0yylr5ysUEAwHWGh5A2S4nq/58HJccODLs2BZ/+yOBt1Q3UZLrJj/LRV2yk8BBJ06H4ONfH9p+E5C/BV67Inq1R11btyEAhpaEd7j+oIRxx9B81vPsNu8Ebn1bpYyorFff/96SiBKiJ9wPe50bvi6G7ToWQqpUEB6nsCJbIzSAAexEmqkwTCcFI2ahyRcg2+b7/5cfK08aX4w0CebEfGuxAmlDdrF61iMDwmqUW/AIUc7cmw9nUl61Wj/yEMhXGQbe7X+5Wmdeh23fq3Tf0LZ+IYNIVBDmRON1pJRylPFq/o2Kd1zaEqYBxBAAgXZoAMP2h/NfgaJh0H+P2K6iac7CjdVh719bZAVjL9iwK+Qv/kPhoXDr9lB66Joqy/8/P9uF2+ng7H2Vi+M5xmuyGoDdc8ik3uunOMdDfjIagOEa2CRduB2C0X3zEWbh87YKgA2fw67oKlE+jxoxugzXwRyPk/V3nxDabo7EV5SrVAQvL1AdjjkPsnJ7PTvqbMJ230vhNFut5ewiVXktCRwhDcBhaWwRGsA7gWm8M97mWX3sPaFKYM2+YJjn1wGjleYaq0iQGTmbNiag1jDTYz9+jDKpRXlaCfounQmPGKavwTZNzszlZZrM/n2Uta0p+h7tCSQyAU1J9NeVjewU7HMA2xfH2G4KgDaOdMb+CH6xAC77pP1t60ZuOCZ8Es3uJ332I1bFM4dDKPvryf8AoKZSjWjPmDIklH1x/IAC1t99AudMGwYkNwcgpQzzHLLb1otz3eRlOalqaGVizrh2PunEaXTQATNQaGUbM5fbbb1m1G7+ALwFKuePqQFEYqbBiJxItf8G322qiZ8fadxxsHNVkiYrpQG4nY6YGsCx3ru5xnc1rzcYabz77Q77W1qN3QQEhIq9x5wEDmkAGSIAzKSNO5bAk8fBSxcaG6TxX+Ba9B9rf1sqD5+h5VFfbgV3mtjcRXsSia7qXxP8/SX1Tetk7BrAxrnR20MCoB2qnsuj7K0ZyJ5D1E2/74gSSnLdlOZZ9ml75sRQv2fMe9RXKw3gtL2jEriGOsn/JJEh1N7pzDhgBH89ywrs6VuQxaod9XyzflfiAB3j2gYRIRt9UBjX8bvn4h0VG3tCNTPIqN9umP2sy5m4I4wUenbt5bKn5zPylndCdWrtLJDjlOki1r0ZgdIAHLidSgPYVtPE+yusEepqOQgfLiWUblgbFccSJQCMOZzPVu1ka3UT79qCwsw5gKSdBbqbyISFy95QJrC3fw0Y97vLltjObS03Zg9SAYRvXmOl7J5yIXgKVLbfHkgiE9DhCf6O6MpGdgqmABiyrxppRWYNDHkBpUnEYxfhsXVo00aWUtukOqxIs4tZQs8MJmqqUR2lXWCYmBrB7BUVrWYEbbDNOWS5Hew1pDj0vm9BVqgO7a5EWRqlbXRntDMgbNcxmUItBk1NKnK45ZyXYcIJMO5Y+NGfQjEAsaKoTXY1tIQlxPto6XZqm33keZwcMcHyVrnhleiI60vnGYI0KU8gYw7ApQrD/+w/C7hljhXx7DcmwJt8AeUxE5GkMNIEZAq1T1dWcMa/vuTKZxeyYMMuRtz8Nl+u2Um220GuJ0Oei367hafoBlU8xmDSwDzL+ycCp8fDEjlCvdm1Rr1OvkC5dzf0MgFgRwixhxDiLCHEheZfqhvW6ZgCYNDegIz2ubbHAfQizKjgoISCbHeosEt1RA7+UGoGw07qrVfJ2sryowWAfRT88vzw2sBvfb+V6karM6/3Wt+T5XJaggbom58VSuiVuCaA6nSVBqBuaWnX5NqQefODH5RH1x4zG5Vd/rwXYeCe+IKtj4QXb6kJi334qVHLYGBxTlxXWS5+lxt8l1NNAbJoKOxY2mobTQ3AY2gAVY0t7KKQHcc+AsfdFzLXDI4TvdvsD8T16zfLP5pJ5mavqGB8/4LQfZIR7H1BKLALgE/vtpaXvxX3MI/TwbygzSS63xUwdJpK+R2Z1bWHkEwuoD8A/2f8HQ7cC5yc4nZ1PmYHP8DIzxL5oLV3DiDDMcPlA0HJ4OIcymubqWpooTaeADBMQL7anThEbA1g4kAravSdxZY5YUt1E1c/962qZWuwaJPVOQ+MKDretyCLohzVkSdM0xsy71kaQFihFG9t1CHxWF+ubL0tuMIEmaUBxH9kWvzBqInv2iY/hdmusLQEdrYV783LgcMACA470MokmgBhCDw1CSxDHjqVw4+H/S5nRJmaR4lMz23S1BIgp5XArkZbJHbapIFoC2Uq6A1PRHoWczL36DvgojfDNnlcDv7uPx3f2c/DH6pVsJkQSqvYuTL1be4GktEAzkSlgiiXUl4M7AVknr+j2UmYBTzqIrIxtscNtAdg9pdSSvYaWkRQwvrKhqgRt9NhpGfILgIEvvqd9C/MDiu/ZyKE4P1rlZfFlGFWHYVKow7Byu2W+e36l600CZMGh99WfQuyQp3PAx8meABNE5CEwhylwTWKfPymGagN6rsHH17pAkRYamBTGLjiTAKDil6OTBdR0+SjKMdNiy0PvT23vukmCtCcO0DVlm7Ve0qCcOA23EBNM55ptjPb2uyPPW8SOQcQC3u70qYYTFs44ndw7osw/jj1Pn8AHGJUbHO44IBfKBdQG26nwIsH76ijwz36PHnhhWh6EMkIgCYpZRDwG8FhOwgv5JIZmHEADrfqxCLdugK9cw5g0uAiTp8ymL+eNZmiHDWaf3XhZuasCM9NM399Fbv9/j0+X1MF7hx8zQ0MKIrfMYzrn49DhJuDdtSqDtVc9/Gy7RHHqNHaq1dO55IDR1KU42ZknzymjyqLG2ymME1ADqaNVC6NwunivlFPqc1tUN89+Gkx8gjZ6xybE+LxvIBAjZojSw7WNvsozHFz7B7KBfPw8X1DQuL1RVv405uWJtrkLFQDlZaI+akIzGygHpeD8trmkEltW00zs5fvCLmkeuPkeGr2B6MCuy7Yf1jY+427rDmF/UdlRqnTMAr6w/hjVZUwUIkadztJLY87NuYh5mBm+p0f86UtzTlOj3InTjCftbainr9/tCrjSkom09vNF0IUA4+hgsDqgdZdFdINUwNwOJXP9DePwLF3WSafXjoH4HI6uP+syQCs3qGCZ575ykrKVprnYVdDCwsMX/0LHv+alYUegn4vxTnxtSUhBHkeV9gkrzmi3lnfwv0frmTZNss0M/v6w0L29X2Gl7LPcKvTGdc/n6XbEphxjGsrsdwVHQ6ochklROtiVMqKQxYtoURy9mynflMDSGAC+mFLDTsjCorUNPkozHZz2t5DOHb3gTzxxTpmr6jggQ9X8vePV4Xt2+AsoA8ol8OEQYVKAJilPbcadvtfPP9t2F6xCvI0+wK0+INhgWAAd5w6iS9WV7JuZ3T21CERQW8ZhZmTKbdMeXVdtxSKoj3XwBIAdV4/v39jCfeduSd7D7Mlgwz641oIzv/312yraeaiA4ZTnBsn0jsNSaYgzFVSymop5cPAj4CLDFNQZmF28MJhRfotsxUgM3P69zITkJ1YZf8evkAFyqy15f3f5RXgaw7Z5+ORm+UMm+S1B0L94+NVfLjU0gBGJsitY05Oxx1dSUsDMAWAy+GgSboBkdAL6OB7P+Gx/1mBX9mOAC3GuMgegWxO7iYyAX21ppIPjHMyg+GqG30MKVFmrByPkxKjc4js/AF2ZRuj8FaSwjmQIEQoU2c8vDFMQGZKCnvFNZPCOEVf0qYYTHswn+fhquJXvM4fwif4V++o57SHvmT2ih2Wa3i8NPJYKbbjzbukK8l6Ae0phDgZmAKMEUKcntpmpQBTAxBO6GukcPDWw/J3YMV7VkHvyMpfvYjSGCMX06Zuxy/cEPC2KgBG980PKxMZr9xerInkyDYEJfFTQoRpAGpk6xDKsylUKCWSgB/8XjbtagplIZVS4pItuD3KtGXPQjpnhcrTE2vOA1TOmbW20XORTZhOH23liEpUW3dbjuHDXrEs9g6+JtiyAAeSFkdO2ERtLGKl+a41zimWV8/ug2NrHYXZGTwoMjtv0XpXF0ujXVlep0xAkDCq3BS2bS2C1N0k4wX0BPAEcAZwkvF3Yorb1fmEBIADLjWSme5aCy+cC8+frQKA3LlWOHgvxBGjU4isBHXpQSNpCrpxBLxqgnbHclX8PMbo6PDx/Vi5vZ7HP18HWHMAkTw5Y9+E7TInIbfXxBvJm6MuEbLROx1CjcZc2VapRDtPHAP3hQcNtQSC5NOE36VMHnYT0COGlhAvDiA/K9ykYvey2X2Q1bHGOnpcf2WmWFsdhPz+qgBNLD67Hx5TIThrcifTmKBOQo7bGVMDGF6mzi1WucSSOMIpozWAIca91S92ynI7U0dEP/ubq5osLSKBBmAO/OPNu6QryVzZ/aWUrf966U7Ixu9QxToGTArPvVK/PSOTuaWaSAEwoDAbL26y8HHkbv3g7XNU/py+EyyPC4NJRpTx7W8t5dKDRlJR18zBY/uEqjKZ7DW0OGEbTH/2zdVNYdkrQ4RMQCLkbukQgoCUSqjHMgFtUfkMlc1fjfC8/iBloo6WLBV8ZjdfmTgi3Dnfv/YQFm6sinJTNb2XfnfixLDRdqz6u6V5Hkb3zWPxllpVSnLlB+qcIl1HbS6iW3LG88DZw/jJ499EnxsqP1OsOQBTa4jl2XPg6D48OHsNL/1sOmP65ROUkj75mRnhHmLsUfDzedERwjEozfNwyLi+/G+l5QDxn6828PNjggyAmBpATZOPu9+1NLYepwEAc4UQmS0AgkFVuAWUCQhUDhB7LEDFCi0AgJuPmxBW/SlSAORlufDipsAVYEy/Aqg1Ar1qoydad7PFA0gp2VHnpX9hNj/eZ0ho/Rc3tx5UXmio5nGTwtkigU1yPU41QnZnJ6wLPFBUMgglkFaW11FGDf4cZbK5853leP2BMJfYpoiUFOMHFHDutGFRfvWnTh7MwxdMYcYBI8LWj+tfwIPnhafSOmRcX8rys6hqbIEhU8FbE9t11Zb7XjpcHDy2L+dOi+2QV5DlijkabWoJ4BCxc/scMKYPi/94DNNGllKa58n8zt+k77ikEzU+eN7eUdesqtkY3gda+HDpdn76lDVwvP+DFTz/jZXaoyWG0E1nkhEAT6OEwAohxPdCiB+EEAmqh6chb10Lr/9cLZteP7l9lM+1yc6VWgAAVxw6mrd/eXDovcvpYP5vrayIeVlOvNJNNqbbrKEeN0ZPKhbluLn6cFV1auQt77Ctppl+BVnc92Mr349ZwjERpp97/IdL2v4rlPeSD1w50T7ctsnkOVm/5svsX0JTFT95eDbDHBW05PYPbV9ZXs9dtkpl8fznTXOLaeN3OwXH7jEwpq19RB/Lq+azGw/nikNGU5rrUQLADFR8JsY0m6HF/tP5k5Cou+v0Pfn2dz+K2jU/2xUzDqCxJUCuJ35gWtoUf+8mCrLd3HzcBO49w5oLXLLdMCEGfFz29Hw+WrYjdL0/WhZewyGW1pXOJCMAHgd+AhyLZf8/KZWN6nTsVYPMtLmRlYT8zZCVJnVP04CXfjadXxp1VO0jwXxDA3CbAsC0r8cJtoq0Q5uxAy9evj+/OGJM3ElVO6Z3RlwBYHTo50wbEVpVmudh2bZa/M4YGkBkrnigZtsa7nU/CkBz/tCQYNpU1RhWM2FyHHOV2carDx/D+rtPiNvBQniKhqGluTgcghJTYA0zvFXKv4cnjw/PWeWtgyH78ozztLABbSzPpDxPbA2gscUfPy2FBlBC/qx9h/L6zw8E4KOV1WpDoCU0B2RqhTvrw+e1Mk0DSEbcV0gp32h9tzQm3xrRhfyCS2OUNIgsq9eLmTaylGm2Yu998j24nQ7ys1xU4WGi/A7+Mh4ajBFQnGyJVx8xlkHFOSzbVsv7S7Zz+HglePcbVcZ+o5KroBYSAPHsq8YEv9tm1jADslZU+ti9XzPMukIJ/xPvZ+f2LURWb77x329yiasagIrhJ/Ha4UM44O5PqGv2MdsIipt7S3xz1WUHj+Lrtbs4cc9BrZ5Pca6H3584kQkDrfmMklw31Y0tSOFAnPkEvHKJsvl/94LKR9N3gjJj5vU1ogCsTj+WEFVzAAGklCFhVNXQwgvzojORamJj1pduNuJCaKknx+2kzuuntslHXbM/asSfaXMAyQiAb4UQzwFvAiFxJ6X8b8pa1dnYcn7jMfzNR8d4mIOJ3ep6M1/dogqZl9c2s928beptVa7iaABFOW4uNiqF3Xtm+767VQ3AMP44ndbI1uwgq31OpQF897zacOL9lG/bTB9gfbA/IxzKb/8i5wcMF9v5OLA3OQXFIRu53ZVyYFH8AcLwsjw+/NWhSZ/TJQeFV08rzfPgD0rqvH4Khx9kbfj8AVXU3WSfGVHzw7E8kwqylOusLyDxuNT2vW//MOn2adRE+e6DClm51ZhnKf+BbM9I6rx+qht9VMbIUNuUwDMrHUnGBJSD6viPJlPdQPvvbi2bo3wzIMThgjGGDbU9BcR7CS6nA5fTwaCiHLwywl0wqyilBTPMOYB49tWgIbg9NgFgdoqN0qPMKTaEoa0sk1b6gwOcSxkgqpAoM5dp6/f6A/TJ93DefuGpEjobM3p0V32LSmPw45lQMjK88wcYdoBKBm3r82PNM+QbrpuxXEE1yeF0CN7+5cH0GWAIgOaa0GR/TZMvFIh31+mTQnMG9d7M+r0TCgAhhBOolFJeHPF3SRe1r3MotKnl9ifnhjVw/SqV9Q9g/yu7tl0ZiMMhOHO/MeErCwdakdQpwBQA972/gje/i/Y2CgRNE5AlAH40UZn9cnIj3C6bqtj9058BsDwY3aln4cPjcoRpAF5/MOUVsfoYabUrGwwle/fTrIFLjpVQj73ONqY8rPtYCBElBMzJXLvQNGMkTp8SPxpWE82QPmpuUPq9IQHwzfpd/NHI43TGlCEcZdxvrZYvTTMS3tVSygBwYBe1JXXES++Q10cFfpWNVulfxxzZpc3KVIQ7woc8v1/sYKtOwuGwOrhfPP9tKC+PSShPj00AHDWxP06HICc3IsXELEvImxpAUFqd54uBwynKcSuNxyFCuXNSXRHLtDdX2KOlD/iler18Dpz5JJz7grFBRnk1RqYgyIshAHI9Li6aPjyU+0mTHKP7FdAinfzrk+VsrVH3+XdGhPuxuw/A43KQZwQC/vHNxGk80o1k5gAWCSHeAF4GQsO8jJoDSIYMLOjebUSWv8wfANu+i71vJ2Hv4JZsraXJF2BgUTbDy/IIBgwTUEQnXZDtwktEWyuWhxa/C6qc8QvkWPYVKt10YOJpoeRnWS4HD81RlaGykvBW6ghleaqdO21pmBm2H9xm1EsoGRG2f6K79f6z9goJTHspTV8gmJTXlSac0X3zacGNB18oPYiZLvu6H40DrBQkGZYKKCkBkA1UAvZZUwlklgA48QHI69f6fprWcUVoAAUDoMXmaumtg1UfwB5npOTrN+5qDGW+fOqSaUwOpWoOd290Ox00OSIyWdZsRiLYvflxGsnmopabWBkcwo+dnzJPjufEsZZ/UIMt105rtYA7Sk5ozqF1L5JEGYfvO3NPTp8yhPeMQjx2V9Cu0GR6IqP75uPDhRvLvGPa/+0utXkeZ9g9kwm0KgAyMvNnLKZm1rRFWhOpAWQVqJrKAZ8yt714Aaydo/KwFHfO5GkouRtW2mqAi574hoWHqYcuyx1u6vM4HdSJiNiOoI+K0n1o3KqE2KdBFZT2j4AKvDo1Tq4fM9tjqjDz8yczaRs5CQzw59P2YFd9Cz+eqiYszRHpJ8u343QInvh8Hf6g1AKgHYzqm0c9Ljw2AWB29Hm2wLlzpw3juW82Rh2fzrQqAIQQQ1DlIM25gM+Aa6SUm+MfpenRuCLcIc3gutotylSxdo56nyB5Vlt55coDeP7rjSzYWMXCjeF5d/xGp5nlCe/ctlQ38UltgFMjko36HG2P97Cnsk4F5iRzMsnEpAyPAwA4f7/h4Z9nCJS/fLCSv3xgVVPTJqC2k5flwlNYgKc6eoI31+OE7UshK59+jmpVFCgQTLnG2Fkk08ongTeAQcbfm8Y6TW/FTK07+Xy4ZYuVcXFTRGIyf+zsn+1hyrAS7vvxXkwaXBSmAYDlOROpAQDsIjp5XIuj7SUOC1KcElkIVeErKRMQrU9ZZbliR/um2pupp+Ku28QZzs843/lRaF1ZnocsRxD+NR3+NonLvzmOS51v84+PVuLdsRqqNsCfymC1dQxf/B3m3B3jG7qHZO6GvlLKJ6WUfuNvJtC3I18qhPixEGKJECIohJjakc/SdAeGLcado0rumQW4m6rDg+kCnScATPKyXGH1agHe/V65hnpidHpV0nADdedBgXIH9gllwopVGP3QcdY80cHGfMDtp+zOrSfs1vHGt0KWyxE2aRsPKRNPApufFQutAXSMP7uf4KZjJwCqwI+oCTeE/M79LLX/e5Csh/aBv++pqoi9ea3auGkefPh7mHNX2gSdJnM3VAohLhBCOI2/C1CTwh1hMXA68L8Ofo6mOzBrK5jdUHaxem2qCg+68nd+YF2u2xkVbi8MgZTtibZoVkmbBtBPPbheQwD0Lwyfy3j8oqlhdY7vPG0Sr145nZ9MH9ElSdKy3c4kJ4FlwlxDEF9D0HMAHWNjzoRQtbcpw0rCvMp8pcoj6Db302pFrkp1Iuu3s3hLDWxZYH1QTXpY0JO5Gy4BzgLKgW3AmUCHJoallMuklHGqXmjSHtMNxTQFOV0qkV7TrnB30BRoALkxOmIhVHsizR63HDeBKgwNYNh+IS+w5ZV++hdmhXzvTQojKkINLc0Nq02carJcjqQngVtj4sDCmOt1Irh28uuVMGw6w3L9lLx6Fp+d1sJ9Q76A589R2y/5gJafzeWzwB4ALA6OYNOM+bDvZUi/j1P/bw4sf8v6vPodUV/RHSTjBbQBOLkL2qLJFMx0Gtm2TianRGkAdgHQiXMAJnkxOrB4GoDH5aCJbGrPf5fCoXvApyrie3uTIK/ERVFO+Oxwd1e+ys9y8d+FW9hjUFFUrqAwYtSKiUQIwbnThvF8hFdKvHTWmlYo6A9FQ2HjXKhczdC1s61t/SfB0GnkAjtRKeVrZB4vz93KH4fuhUNIlWl2/efWMQ0VpANx73ghxO8THCellLcn+mAhxEegCulEcKuU8vUk24cQ4nLgcoBhw1Kbj0WTJHtfoJK/Tf+5tc4UAE1VSjOQwdQIAJsGUJDtoq7Zj0ASlCLK7m2aO5r6T6EwOzsUv+DHyRlThrDeqOE7um8eayoawtI0dwdDS3NZXl7Hn95amlAASIjyAoqFmSLDTqx5D02SxMoWPOFEOOdZQBlERw3qB9uhhjwWba6Bgw4A4HSn0fmPPBTWfarqPk84vosaHp9EJqCGGH8AlwI3tfbBUsqjpJR7xPhLuvM3PudRKeVUKeXUvn07NPes6Sycbjj0BvDYgqxySlTw16oPrDmCFJiA9rXVbX35iun8+bQ9EKhykJGjW3dkIRlj2CxQ5hazGP2Z+wxl7Z3Hp9zTpzWGl+a2vhPmHEDr+11z1Lgom3+ONgG1H3uyyD7K3s8J94ftstfIgQAUFvfhu03VbHYM5F7f2dYOQ/dTx26Ym+rWJkVcASCl/Kv5BzyKygp6MfACECOZvqZXY09YZhKjTGRHGT/AnkPfw6g++TgIIonWALIi6wgYBX8aZBZZbmcoA2d1YwuOOAFgXYl9AnrZttq4+ykNoHUGF+fwxEX7hq2LzBmkaQOmtxvAjLfh/FeVachOnUqR3lCousgfPzyXZwO2HGPZhVA21poE3rUOVr6fylYnpLVsoKVCiDuA71HmoilSypuklB2awRBCnCaE2AxMB94WQnTfL6DpHOwC4KK3oN/ElN3YpqmmX0EWuw0siKsBRJWSnHY5Nftdz1OBY8hyOThjn8GM658fFUTVXew30iqQc9zfP4va3uIPsq2mKWa9+HgcOKaMe8/Yk1MmKxdYPQfQAQ68zlrO76cKzkdStQ6AJU7lcbatppk6bJpdViEUD4WaTcqZ4sFp8NxZqWx1QhLNAdyHctV8FJgkpYyuo9dOpJSzgFmd9XmaNMAUAMXDYeTBqorVsrcSH9MWareG0nq/fvWBNHj9CKE6fYcxJo4sjBJVSMadzZWbf4SXSrJcDvoVZPPBdckXcUk1k4YUMbZfPqt2xH7UTnvoC5ZsrSXb7WjVDdRECMFZ+w7l5MmDOGHSwLglLTVJ4HQpT7LIUb+dPc+Brd9y5rFH8uqO5WypbiJoH2fn91e5slrqlRAwzUotjape+cMHqSywU36S2nMxSKQB/BoV+ftbYKsQotb4qxNCxNdPNb2TnGL1Ov449ZrfXxWKD3RCfvRt38P9u8GCmYCqUTy8TKV5VmYeaUTHhneK5hyAzxY38OUaFcKSrsW7zaLykfgCQZZsVY9dMIlAsEiy3U6O3j2WT4amTfxqKVw2J/72/a+A3+9i2KBBPDFj3+jt+X2t+YOF/7HWN+1SkfQ7V8IbV3dqkxORaA7AIaXMkVIWSCkLbX8FUkpdPV0Tzp5nwxG/g8NuVu/z+wESajsh4MW0l8bQKIQQygsoxq2cqJRkVYxyfunAroh2zVmxg2/W7aLJFiGs7fjdiNOtNIFEOJSZzT5fdaL3Dir6HcDUx7ZSM9Ao+blrjXXM1kWwzCi93kkJFJNBhwVqOof8fnDI9ZYpaORh6nXpGx3/bOOBork69mZDA4jEFADeGIW69xkeY9I6DaiyZR2VUjLjyXmc9chcmlsiBED3z1lr2sBiOYpTa29gZ7ODH8obwZOvJoBNvn8Bvn9JLXeG1pwkWgBoUkOfMSoz6NZvO/5Z3jr16o+dkVMgkTF6RHMSeOnWWp75agOgJo7PmDKEqTZ30nTCrgHc/tay0PLPnlkQtl8ycQCa7ufxi6YyokxNAm8xagg4BCp9yq61aqeyMbDsTTXAySpUsTRdhBYAmtSR3x+MAuwdwmtMOcUJLHPEEwAuq5bwb19bjC8QpMkX6PaI32R54gtrhPjtxuqwbbqAXWZw5G79efTC8HyX/qBUmrKp0Q62be83EfxNiav+dCJaAGhSR24faOho3kBUWl1QmRVj8ON9BpMXI4jLE5H5sqqxhWZfQAdDabqUgUXh6ccbvH7LaQKgz1hr2bT/x7nXO5vMGAppMpOiwao4jFkprL2YGRebYzufFWS5rMR0NiKjYLdVN+MLyC7J7Nle3vnlwazYXsvK7fX8a86auPs1ZVjpwd5MZIR5ndcP2UXWirIx1nK+kY7c7+3YM5MkWgPQpI5Be4OvwRrBg4qU3Lkanjweyn9Q69Z+Cu/eHPsztn1vdfyNlbHzqMtgTJtIZO77tTuVf31Jridq33Rh4qBCTtt7CCU2d9Dx/aOL2tR7u26iUNNxbj1+N647Srl/7qz3WmZNgBEHW8v5RoxBoGu81LQA0KQOc5TTYgts+ut4+Oc+sOELeMYoGv/0yfD1v9Sop6UBvMb+C56CRw6GjV8aB0toronxRbGTI0RqAGYlsZI4vvbphN0b6LDx0Tmw6pu1AMgkLjtkFNccNZaCbBfba5rhwGvVhvNehrwyGLq/Mv94VHxLZ5ZTTUT66sKazMe8mVsaYm+v3x7+vmEnPHa4ev3DLvjfX6KPiTUykjKmCSgyN9D6nY0AFOWkvwCobVIdwG0nTQwTBiZ13q7pIDSdy4DCbMprm2HMkXCbbTAz421l9//hZfU+BYkUY6E1AE3q8BimC1MDiCUIfDbXzqdOVEJBBtR6afPfdxrFW2KNjJI0Ab39wzaAjCjYfeVho9l/VCmn7j04VODdTmE3Zy7VtI8BRdm8v2Q70/78UXgwotMF7mxwJbjPU0D6PwmazMXUAJqq1avpzz/2aJUzBZRd38T0iwZYOzs8iticKIvUADZ8CQuejKkBOB0CZ4wsn7HWpRtDSnJ54fLpFOd6yI6odFaW5+FPp+zRTS3TdIR+BcojaEedl3nrd0XvYE78+r3gS707qBYAmtSRawRbzbpcvZqBXBNPhbE/Usubvo597Pu/CX8/+nD1Guke96SReyjSnGQQ6QoKmSEA7ETOZRy9e/+ocpaazKCxxbp/Y+ajMjXdf02H+8bC+7emtD1aAGhSh+nSZmIGcrmyLD/orQvV6+QL1OvgfdSrqQ1cORcO+AUMMRJrRarGrsRVvGIVQXdmWBRVZC6jXI+eustU7GbJijovd76zjLlrbFqwy+ah1lIHS1KbNFkLAE1q2f8qcBumIFMDcGWrUHiw8qEceiPMeAcutBWMm3gK9J8IR99h2UaDEQJg6DT1aj/ORuQ8AIAjw+76SJdPXdg9c/ntCbtx0XRVf+LB2at59H9rOfexr6wdikeEH9BYmVIzUIY9CpqMw51jeTSENACbAKg0gp3y+8GIAyGrQIXDAwyYZH2Ow7CNRmoADqfSDkYdFvPrIz2BIPNMQObz/7NDR7H3sGJ+sn96FLDRtJ1+hdn88ZQ98DgdVMbKSFs6CvY4U/0ddot6duJ50XUCWpfUpBZXtrLbB/w2DSDLihGoWKaW7QW3xxwJO5aqQDITMwXv4z+CP1RbXj9+b6jYeyxME5DH6QiVhsw0E9BPDx5Jsz/ANUeO1RW9egj52a6wxH+qzrNQ6umZj6uV859Ur95a8DVGm1Q7Aa0BaFKLabrxN4drALllKlsoWHZ/k0NvVn7Ro46w1jlsbo81m6xlf7P1HTEIGsPnoaWWgMk0DSAvy8VNx07QnX8PIi8r/Fp6/UF8gSDvLylHmipfluFGveFLFUC5/O1Ob4cWAJrUYo7OK5bDineNdVlqpGOO8McfH35MVj6MOCjcWO+0TY5V2wVAYg3AfJbswV+ZJgA0PY8JA8JrajW2BPjDG0v42X8WMG+9kQ46y9hn8X9VrMvQ/Tu9HdoEpEkt5uj830da6woGqtcjf69yn+yZRFFsexWmOhXQxbrPYPtiKBoa97C+BVls3NVIni0BnCPDTECansduAwr4cKnluvzvz9by3NcbAVi3s56hpTkMNGpgs+JtFVSZV9bp7dAagCa1xHLTzDdy25SOguPuCc+MGA93rrXcXA2121TkMIRHDEcwtl8+ED4ZrDUATXdTlh9utnzIlvn1pld/YPpdn7CwZbC1gzuxu3N70QJAk1oi7fM3rI29X2v0GQcn/UMtN1XD61dZ2055MO5hZqF4l0MLAE36MKAovtnS5McPz1Vu1BAz0r0z0AJAk1rs9vnbatqvxgoB+1ykPq+5BloarW350dkyTcy+3m3TALQJSNPdHDK2L0ft1p97z9gz7j6BoFRzYaDmxVKAFgCa1JLAQ6dd5PeD6o2WG6jdNBQDc7SfbRMALq0BaLqZHI+Tf180lRP3Gph4x2Ij5mPs0SlphxYAmtSSwEOnXQyeaqSPMDpxZ+LiLuZo355R06EFgCZNyPW4uPrwMfF3GLAHXL1ARcOnAC0ANKnF3ckCoHiY0gCkURmslbJ5U0eUALD30JLQOj0HoEknXE7rfsyJFevRZ4yKeE/Fd6fkUzUaE1MD6CxNoMRQic0sovkDEu5+4p6DmDKsJOzByrRIYE3PpslnlTktzHGFvW/2BVIaAKg1AE1qcRhjjLxOCmM3s4aanPLPVg8ZVJxDji2BWqYlg9P0bC47eBRj+uXzyhXTo7LX1jantjCMfhQ0qSXHML1Mu6xzPs/lsSbG+k6AQZOTOiwsDkBrAJo0ok9+Fh/96lCmjiglyyj+c+KeanK4tim1tZ8z3gTk8/nYvHkzzc3Nre+sCSM7O5shQ4bgdqewvGBeH/jN1la9ddqEOfHbBrOSEIJJg4v4YUuNngPQpC1XHTaav36wkuP2GMhb329LuQaQ8QJg8+bNFBQUMGLEiFDQj6Z1pJRUVlayefNmRo4cmdovM0tDdhamAGijUHnmp/uxfmeDvk80acvpU4Zw+pQhLNig8gHVNmkTUEKam5spKyvTD3UbEUJQVlaWmZqT6fnTxvD4ohw3ew0t7vz2aDSdTFGOGpvXNqfWBJTxAgDQnX87ydjfLaQBpCY/ikbT3RRmq0FOj9QAhBD3CSGWCyG+F0LMEkIUd0c7NBmKFgCaHk6hkb68p3oBfQjsIaXcE1gJ3NJN7egUNm/ezCmnnMLYsWMZNWoUV199NV6vKn7y+eefM23aNCZMmMD48eN56KGHwo599NFHmTBhAhMmTGDq1KnMmTOnG84gw2inCUijyRSyXA48TkfKvYC6RQBIKT+QUppn9hUwpDva0RlIKTn99NM59dRTWbVqFatWraKpqYkbb7yR8vJyzjvvPB5++GGWL1/OF198weOPP86sWbMAeOutt3jkkUf4/PPPWb58OY8++igXXHABW7Zs6eazSnPaOQms0WQKQghys5xsqmpsfecOkA5eQJcAL8bbKIS4HLgcYNiwYQk/6I9vLmHp1tpObdzEQYX84aTd427/5JNPyM7O5uKLLwbA6XTywAMPMHz4cBwOBzNmzGDKlCkA9OnTh3vvvZff/e53nHbaadxzzz3cd9999OnTB4ApU6Zw8cUX8+CDD3LnnXd26nn0KFrq1Wtn5xnSaNKI6kYfb3+/jX+cI1PmupwyDUAI8ZEQYnGMv1Ns+9wK+IFn432OlPJRKeVUKeXUvn3jp/3tLpYsWcI++4TXtC0sLGTEiBF89913UdumTp3K0qVL4x5r366Jw4Yv1Gvjzu5th0bTBTS0pM4MlDINQEp5VKLtQogZwInAkTJUBbljJBqpa3oQB/wSvvwH+DLQhVWjSZIhJTlsrmqivtkf8grqbLrLC+hY4EbgZCllao1cKWbixIksWLAgbF1tbS3l5eXsvffeUdsWLFjA1KlT4x5r366Jw2BlUiPg7d52aDQp5ObjJgBQ702dBtBdXkD/BAqAD4UQi4QQD3dTOzrMkUceSWNjI08//TQAgUCAX//611x99dVcf/31zJw5k0WLFgFQWVnJrbfeyu9+9zsAbrzxRm666SYqKysBWLRoEbNmzeJnP/tZt5xLxuA2I4szNI5Bo0mC/CxloKlLYTBYt0wCSykTVEDILIQQzJo1i5///OfcfvvtVFRUcPbZZ3PrrbcC8Mwzz3D55ZdTU1PD+vXrmTlzJoceeigAJ598Mlu3buXAAw/E7/dTXl7Od999RzrOdaQVo49QtVIP+GV3t0SjSRmmAGjogRpAj2Lo0KG88cYbrFq1infeeYf33nuPhQsXAnDIIYfwzTffsGLFCh544AHuvPNOqqqqQsdeccUVLF++nOXLl3Paaafx+9//nk6aEum5OF1w7F1Q2Eo5PY0mgzHrADTb6gN0NungBtqjOOCAA9iwYUPMbVdddRVXXXVVzG0ul4v//Oc/qWyaRqPJILKNMqZNKRQAWgPQaDSaNMSsDeD1BVP2HVoAaDQaTRpiVrFr9msNQKPRaHoVXTEHoAWARqPRpCHZRhnTZm0C0mg0mt6Fy+mgJNfNhsrUxcpqAdAJOJ1OJk+ezF577cWUKVP48ssv+eGHH5g8eTKTJ0+mtLSUkSNHMnnyZI466ih8Ph8333wzY8eOZcqUKUyfPp1333036nNnzpzJ1VdfHbbusMMOY/78+QA88cQTTJo0iT333JM99tiD119/HYAZM2YwcuRI9tprL8aNG8eFF17I5s2bU/9DaDSaTuXAMX3436qKlLmGazfQTiAnJycU7fv+++9zyy238Omnn4bWzZgxgxNPPJEzzzwTgJtvvplt27axePFisrKy2L59O59++mmbvnPz5s38+c9/ZuHChRQVFVFfX09FRUVo+3333ceZZ56JlJK//e1vHHHEESxevBiPx9Mp56zRaFLP3sNKeOv7bZz1yFz+79wpDCjq3Ay4PUsAvHszlP/QuZ85YBIcd3fSu9fW1lJSUhJ3e2NjI4899hjr1q0jKysLgP79+3PWWWe1qVk7duygoKCA/Px8APLz80PLdoQQXHfddcyaNYt3332XU045JWofjUaTngw0Ovx566tYub1OC4B0pKmpicmTJ9Pc3My2bdv45JNP4u67evVqhg0bRmFhYYe+c6+99qJ///6MHDmSI488ktNPP52TTjop7v5Tpkxh+fLlWgBoNBmEvcM/ZFznp4jpWQKgDSP1zsRuApo7dy4XXnghixcv7nDR9XjHCyFwOp289957zJs3j48//pjrrruOBQsWcNttt8U8RqeX0GgyjwGFqS16pCeBO5np06ezc+fOMHu8nTFjxrBx40Zqa6Mrl82aNSs0cTx//nzKysrC8gYB7Nq1K1RBTAjBtGnTuOWWW3jhhRd49dVX47br22+/ZbfdduvAmWk0mq6mX0FWSj9fC4BOZvny5QQCAcrKymJuz83N5dJLL+Waa66hpaUFgIqKCl5++WVOO+00Fi1axKJFi5g6dSr77rsvX3zxBeXl5QDMnz8fr9fL0KFD2bp1ayjhHKhU0sOHD4/6Pikl//jHP9i2bRvHHntsCs5Yo9GkCpcztV10zzIBdRPmHACoDvepp57C6XTG3f+OO+7gt7/9LRMnTiQ7O5u8vDz+9Kc/Re3Xv39//v73v3P88ccTDAbJz8/n+eefx+Fw4PP5uP7669m6dSvZ2dn07duXhx+2yirccMMN3H777TQ2NrL//vsze/Zs7QGk0WjCEJlkG546dao0feBNli1bpk0bHUD/fhpNevPR0u20BIIcP6n96c+FEAuklFGlBrUGoNFoNGnMURP7p+yz9RyARqPR9FJ6hADIJDNWOqF/N42md5PxAiA7O5vKykrdmbURKSWVlZVkZ6fWz1ij0aQvGT8HMGTIEDZv3hzX714Tn+zsbIYMGdLdzdBoNN1ExgsAt9vNyJEju7sZGo1Gk3FkvAlIo9FoNO1DCwCNRqPppWgBoNFoNL2UjIoEFkJUABvaeXgfYGcnNqc76Snn0lPOA3rOufSU84Cecy6dcR7DpZRR+aQzSgB0BCHE/Fih0JlITzmXnnIe0HPOpaecB/Scc0nleWgTkEaj0fRStADQaDSaXkpvEgCPdncDOpGeci495Tyg55xLTzkP6DnnkrLz6DVzABqNRqMJpzdpABqNRqOxoQWARqPR9FJ6hQAQQhwrhFghhFgthLi5u9uTCCHEUCHEbCHEUiHEEiHENcb6UiHEh0KIVcZribFeCCH+YZzb90KIKd17BuEIIZxCiG+FEG8Z70cKIb422vuiEMJjrM8y3q82to/o1oZHIIQoFkK8IoRYLoRYJoSYnonXRAhxnXFfLRZCPC+EyM6UayKEeEIIsUMIsdi2rs3XQAhxkbH/KiHERWl0LvcZ99f3QohZQohi27ZbjHNZIYQ4xra+Y32blLJH/wFOYA0wCvAA3wETu7tdCdo7EJhiLBcAK4GJwL3Azcb6m4F7jOXjgXcBAewPfN3d5xBxPr8CngPeMt6/BJxjLD8MXGksXwU8bCyfA7zY3W2POI+ngJ8ayx6gONOuCTAYWAfk2K7FjEy5JsAhwBRgsW1dm64BUAqsNV5LjOWSNDmXowGXsXyP7VwmGv1WFjDS6M+cndG3dftN2QU/9HTgfdv7W4BburtdbWj/68CPgBXAQGPdQGCFsfwIcK5t/9B+3f0HDAE+Bo4A3jIexp22mzx0bYD3genGssvYT3T3ORjtKTI6ThGxPqOuiSEANhmdn8u4Jsdk0jUBRkR0mm26BsC5wCO29WH7dee5RGw7DXjWWA7rs8zr0hl9W28wAZk3vclmY13aY6jcewNfA/2llNuMTeWAWSg0nc/vb8CNQNB4XwZUSyn9xnt7W0PnYWyvMfZPB0YCFcCThjnr30KIPDLsmkgptwB/ATYC21C/8QIy85qYtPUapOW1icElKA0GUnguvUEAZCRCiHzgVeBaKWWtfZtU4j6t/XeFECcCO6SUC7q7LZ2AC6Wu/0tKuTfQgDI3hMiQa1ICnIISaIOAPODYbm1UJ5IJ1yAZhBC3An7g2VR/V28QAFuAobb3Q4x1aYsQwo3q/J+VUv7XWL1dCDHQ2D4Q2GGsT9fzOxA4WQixHngBZQb6O1AshDALEdnbGjoPY3sRUNmVDU7AZmCzlPJr4/0rKIGQadfkKGCdlLJCSukD/ou6Tpl4TUzaeg3S9doAIISYAZwInG8INEjhufQGATAPGGt4OnhQk1lvdHOb4iKEEMDjwDIp5f22TW8ApsfCRai5AXP9hYbXw/5AjU0l7jaklLdIKYdIKUegfvNPpJTnA7OBM43dIs/DPL8zjf3TYjQnpSwHNgkhxhurjgSWkmHXBGX62V8IkWvcZ+Z5ZNw1sdHWa/A+cLQQosTQiI421nU7QohjUSbTk6WUjbZNbwDnGF5ZI4GxwDd0Rt/WnRM6XTjZcjzKm2YNcGt3t6eVth6EUmO/BxYZf8ejbK8fA6uAj4BSY38BPGic2w/A1O4+hxjndBiWF9Ao4+ZdDbwMZBnrs433q43to7q73RHnMBmYb1yX11AeJBl3TYA/AsuBxcB/UJ4lGXFNgOdRcxc+lFZ2aXuuAcq+vtr4uziNzmU1yqZvPvcP2/a/1TiXFcBxtvUd6tt0KgiNRqPppfQGE5BGo9FoYqAFgEaj0fRStADQaDSaXooWABqNRtNL0QJAo9FoeilaAGg0SSKEOLldGRet468VQuR2Zps0mo6g3UA1mi7CiIqeKqXc2d1t0WhAawAaDaAS7xm52GcKIVYKIZ4VQhwlhPjCyBs/TQgxQwjxT2P/mUa++S+FEGuFEGca6w8TRu0D4/0/jeN+icq/M1sIMdvYdrQQYq4QYqEQ4mUj/5NG02VoAaDRWIwB/gpMMP7OQ0VmXw/8Jsb+A43tJwJ3J/pgKeU/gK3A4VLKw4UQfYDfAkdJKaegoox/1UnnodEkhav1XTSaXsM6KeUPAEKIJcDHUkophPgBlbs9kteklEFgqRCif4ztidgfVejjC5WWBw8wt90t12jagRYAGo2F17YctL0PEvtZse8vjFc/4Zp1dpzvEsCHUspz29FOjaZT0CYgjaZz2QBMNDI3FqMybprUocp8AnwFHCiEGAMghMgTQozr0pZqej1aA9BoOhEp5SYhxEuobJvrgG9tmx8F3hNCbDXmAWYAzwshsoztv0VldtRougTtBqrRaDS9FG0C0mg0ml6KFgAajUbTS9ECQKPRaHopWgBoNBpNL0ULAI1Go+mlaAGg0Wg0vRQtADQajaaX8v8pZCNKUAucuQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(0, 946.4194546463453)\n",
      "[*********************100%***********************]  1 of 1 completed\n",
      "[*********************100%***********************]  1 of 1 completed\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0       0.527238\n",
      "1       0.638272\n",
      "2       0.416230\n",
      "3       0.813154\n",
      "4       0.935293\n",
      "          ...   \n",
      "1201   -2.401316\n",
      "1202   -2.445731\n",
      "1203   -2.499304\n",
      "1204   -2.362458\n",
      "1205   -2.359680\n",
      "Name: Close_a, Length: 1206, dtype: float64 0      -0.642669\n",
      "1      -0.705889\n",
      "2      -0.737525\n",
      "3      -0.817434\n",
      "4      -0.768994\n",
      "          ...   \n",
      "1201   -1.695561\n",
      "1202   -1.693479\n",
      "1203   -1.648346\n",
      "1204   -1.573819\n",
      "1205   -1.556848\n",
      "Name: Close_b, Length: 1206, dtype: float64\n",
      "(0, 46216.12912316974, [(0.003312828682910043, 0), (0.010361012023485827, 15), (0.027814768926969637, 30), (0.022865170454147123, 45), (0.008536417029482095, 60), (0.008075500805114115, 75), (0.010451413964966984, 90), (0.019641892144205225, 105), (0.009472031991638732, 120), (0.003665410210419343, 135), (0.008744879699222606, 150), (0.014656485337404171, 165), (0.054979393520618564, 180), (0.0002175704849556731, 195), (0.0007731330676298692, 210), (0.03842025890947595, 225), (0.01989292295326676, 240), (0.0003647251834550391, 255), (0.013885955441682907, 270), (0.013791794098319023, 285), (0.020541302342006688, 300), (0.047048304868519214, 315), (0.03287779813465843, 330), (0.0009189630083463254, 345), (0.01286867161818751, 360), (0.005050821659427345, 375), (0.013114394246981744, 390), (0.008018794844573815, 405), (0.02208435209776543, 420), (0.008413157240326902, 435), (0.04605647126099074, 450), (0.050371437643642006, 465), (0.014922343083893654, 480), (0.015297467875301313, 495), (0.02473617942348555, 510), (0.004427322587118919, 525), (0.024517634205052628, 540), (0.005479128169738563, 555), (0.007105529710965213, 570), (0.007939209416075324, 585), (0.0002511208760096143, 600), (0.0038056653633567364, 615), (0.0027635913902583666, 630), (0.00043886072118177327, 645), (0.015580585169846138, 660), (0.0012031873834509443, 675), (0.00017159046951145673, 690), (0.002320095781189583, 705), (0.019361970977113306, 720), (0.020647190960588327, 735), (0.025889229845146872, 750), (0.01701114487746257, 765), (0.02319806392225891, 780), (0.01612155032266568, 795), (0.02307088670025224, 810), (0.012654261405217686, 825), (0.013653857082472366, 840), (0.005698718326254248, 855), (0.00794954958442236, 870), (0.012775961235214033, 885), (0.02385857377189017, 900), (0.00508874922625669, 915), (0.0064304926439538255, 930), (0.03803396951353835, 945), (0.009411265524962825, 960), (0.032344978962566814, 975), (0.00903878993207546, 990), (0.0021572959914183476, 1005), (0.0008119676373772294, 1020), (0.02316394580852251, 1035), (0.010384015663285923, 1050), (0.022276211210064043, 1065), (0.006506206133940677, 1080), (0.006666439885243885, 1095), (0.017985433183299458, 1110), (0.010244693023390797, 1125), (0.00816788922726695, 1140), (0.008501809747237557, 1155), (0.001770511902076945, 1170)])\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAEICAYAAABCnX+uAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjUuMSwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/YYfK9AAAACXBIWXMAAAsTAAALEwEAmpwYAACMNElEQVR4nOydd3gcV7m43zPbd6VV77Itudtxt9N7AZIQWi4QEkII5dJ7aLnAvVxCqJdLu/CD0EKHkBBKIISQ4vS4xXbs2Ja71XvbXW2bOb8/zmyTVs2WLK087/Po0ezMmZkzO7PffOc7XxFSSiwsLCwschdtpjtgYWFhYXFqWILcwsLCIsexBLmFhYVFjmMJcgsLC4scxxLkFhYWFjmOJcgtLCwschxLkE8BQohbhRBPjbH9QSHEW9M+f1EI0SWEaJui80shxOIpOtZ8IURACGGbiuPNNOPdm2k872VCiKZpOO7nhRAx8x75TvIYe4UQl01tz6aGqXyWsxz7USFEeCaeh+nGEuQTRAhxkRDiGSFEvxCiRwjxtBDi7InsK6W8Rkr5c/M484HbgJVSysqZEjSjIaU8IaXMk1LqM92X2YIQos4UMPaZ7ovJ7817FDyZnaWUZ0kpHz/VTkzXy2qC5570PZFSXgG8Zxq7NWPMlgdzViOE8AMPAO8F7gGcwMVA5CQONx/ollJ2TF0PpwYhhF1KGZ/pfpwsud5/C4uTxdLIJ8ZSACnlb6WUupRySEr5Tynl7vRGQoj/EUL0CiGOCiGuSVv/uBDinUKIq4CHgWpzaPx74AfA+ebnPrO9yzzWCSFEuxDiB0IIT9rxPiGEaBVCtAgh3j5Wx81zf1kIsUUIMSCE+LMQotjcltBq3iGEOAE8OlzTEUIUCyF+Zp6rVwjxp7RjXyeE2CmE6DNHK2vG6McFQoit5ohmqxDiAnP9DUKIbcPaflQI8ZfxvouERiiE+JRppvrZWN+Fuc+3hRCN5nexXQhxcdq2c4QQ28xt7UKI/zU3PWH+7zPv0/lZjusRQtxtfkcvAWcP2/5pIcRhIcSgEOIlIcTrzPVOc4S3Oq1tuRAiJIQoG+96zPZ3CyG+L5QJL2COFiuFEN8y+7NfCLE+rf0x81lMmGruEUL8wuzbXiHEprS2GaYO81xfFMqs8yCpZzkghKgWQmhp19ptHrt4jL6P+iwLIV4phHjBvB+NQojPp20ecU+EEIuEMp90C2W6/LUQonAi32HOI6W0/sb5A/xAN/Bz4BqgaNj2W4EY8O+ADaW5twDC3P448E5z+TKgadi+Tw073jeBvwDFQD7wV+DL5rargXZgFeADfgNIYPEofX8caE5rfx/wK3NbnbnvL8xtnrR1drPN34DfA0WAA7jUXL8e6ADONa/5rcAxwJWlD8VAL/AW1CjwRvNzCeAFBoElae23Am+awHdxGRAHvgq4AE+Wc2d8v8DN5nntKBNXG+A2tz0LvMVczgPOG/Y92cd4Rr4CPGn2cx6wZ9h9fgNQjVKebgCCQJW57fvAV9Pafhj46yjn+Xzi/qWtuxvoAjYCbuBR4Chwi3lvvgg8ltb+GHBV2vHCwLVm2y8Dz6W1zXi2zHN9MduznNb354Ba8578EPjtKNcy5rNsHn+1+Z2tMdu+drR7AiwGXmaetwwl7L811vMwV/5mvAO58gesMB/iJpTw+AtQkfZwHEpr6zUfskrz8+NMUJADwvyRL0pbdz5w1Fz+KfCVtG1Lh//YhvX78WHtVwJR80eb+DEsTNue/IEAVYDBsBeX2e7/AXcMW3cAU9APW/8WYMuwdc8Ct5rLvwL+01xeghLs3gl8F5eZ1+Ie476N+cNFvVDWmstPAP8NlA5rM0JoZDnOEeDqtM/vYpiQG9Z+J/Aac/lc4ASpF/824I2j7Pd5sgvyH6V9/iCwL+3zaqAv7fMxMgX5v4Y9H0NpnycryPcBV6Z9rkIpOSO+Oyb/LH8L+OYk7slrgRcm8zzk6p9lWpkgUsp9UspbpZS1KA2iGvVgJWhLaxsyF/NO4lRlKCG23TRZ9AH/MNdjnrcxrf3xCRxzeHsHUDrK9nTmAT1Syt4s2xYAtyX6aPZzntm/4VRn6edxoMZc/g1KSwe4CfiT+R2O910AdEopw6P0fwRCiI8LIfaZJp4+oIDUd/EOlDDZb5p/rpvocRnnvgghbkkzQ/WhnqFSACnl80AIuEwIsRylWf5lEucGpa0mGMryeaxnMd17KgS4xclP7C4A7k+7zn2ADlRkaTved3auEOIxIUSnEKIfNVGZ/twyrH2FEOJ3QohmIcQASkEYtf1cwhLkJ4GUcj9KM1k1FYcb9rkL9cM7S0pZaP4VSCkTP8RWlMBMMH8C5xjePmaeZ7Q+JGgEikexMzYCd6b1sVBK6ZVS/jZL2xbUDzyd+SiTD6h5gzIhxDqUQP+NuX6872Ksvo/AtId/EngjapRRCPSjNH+klAellDcC5Shzzb2mLXgi5xj1vgghFgA/Aj4AlJjn3ZM4r8nPUWaftwD3TublNM2EUC/TBJVpy9m+l0bgmmHPhVtK2Zyl7XjP8m9QL7R5UsoC1HxS4jvLdu4vmetXSyn9qO9TZGk357AE+QQQQiwXQtwmhKg1P89DCZznpuDw7UCtEMIJIKU0UD/6bwohys3z1QghXmG2vwe4VQixUgjhBf5rAue4Oa39F1CCYlz3QillK2pC6/tCiCIhhEMIcYm5+UfAe0ytSQghfObkVH6WQ/0dWCqEuEkIYRdC3IAawj9gnicG/AH4OsrG/PAEv4vJko8yi3UCdiHEf6LmPzCPfbMQosw8b5+52jDbG8DCMY59D3C7+T3VoswbCRIvg07zPG9jpBLwK+B1KOHzi5O6uulhJ3CTEMImhLgauDRtWztQIoQoSFv3A+BO8+WFEKJMCPGaUY493rOcjxoRhoUQ56BGawmy3ZN8IAD0CyFqgE9M5kJzGUuQT4xBlB3zeSFEECXA96Amy06VR4G9QJsQIqElfwo4BDxnDhH/BSwDkFI+iDLpPGq2eXQC5/glagTRhpoM+9Ak+vcWlAa/HzW5+RGzH9tQk7v/h7IzH0LZH0cgpewGrkN9X90orfg6KWX6qOA3wFXAH2SmC+Go38VJ8BDKNNOAGsaHyRzaXw3sFUIEgG+jJlyHTDPPncDTpsngvCzH/m/zmEeBf6K+88T1vwR8AzUv0I6yWT+dvrOUshHYgRL4T57k9U0HHwZehXqxvRn4U2KDOTL9LXDE/F6qUd/bX4B/CiEGUb+Vc7MdeALP8vuAL5jH+U+U4E/sm+2e/DewATXK+hvwx1O47pwiMbliMUcRQjyOmhz78Uz3xWJshBA/BVqklJ8do81ngdtRL9caeZJBQWciQoiHgfNQE+9XznR/phJLkM9xLEGeGwgh6lBmjPVSyqMz2xuLXMMyrVhYzDBCiDtQprqvW0Lc4mSwNHILCwuLHMfSyC0sLCxynBlJmlVaWirr6upm4tQWFhYWOcv27du7pJQjcvDMiCCvq6tj27Zt4ze0sLCwsEgihMgayW2ZViwsLCxyHEuQW1hYWOQ4liC3sLCwyHEsQW5hYWGR41iC3MLCwiLHsQS5hYWFRY5jCXILCwuLHMcS5LOMe7Y1cs/W0Qr2WFhYWIzEEuSzjL/sbOE3W07MdDcsLCxyCEuQzzIq/G7aB2ZLlS8LC4tcwBLks4zKAhcdgxF0w8pKaZH7tLb+kUceXUQs1jfTXZnTWIJ8llHpd6Mbku5AZKa7YmFxyhw+8g0A+vtfmOGezG0sQT7LqPC7AWgbCKNLSUPHc0SjXePsZWExO3E6SwEIR1pnuCdzmxnJfmgxOpV+JwUEaOsPo+cJeva8mePOOl520SMz3TULiwmz5a9HCPZF8K/0AFjKyDRjaeSzjOX/uJFvO75H+0CYzW0vAqBFj81spywsJkk4FOfAlnZ0PQpAQ8N2tm7dOsO9mrtYgnyWYZ+3kYL8IVp6u2ns2qLW2f0z3CsLi7GJxwfZtfvdhMPKhFK/thQ9ZhAeCgAQCHTR2GjFR0wXliCfZRxYeB3XnP0jtkeauTb6PQAcjuIZ7pWFxdh0dDxEV9e/OHLkfwEor7fh9NiJRYIA2Ow6Q0NDM9nFOY0lyGcZP4hXAxAyQhjm7Skuvmgmu2RhkeTeth5etvUAsVgMXdeT6x2OQgAi0U4GBvby1NMbKNz4R7C3ARDSe2nubZ6JLp8RnLIgF0LME0I8JoR4SQixVwjx4ano2JnK7oAKBup3uTkh5wMghPW+tZgdfGDfCV4MDPHHLdv57ne/SygUMrcYAOjxAHuObQaguPqvyf2cDnDqztPd3TOGqZAQceA2KeVK4Dzg/UKIlVNw3DOSfUElyA/66viM9g1CeKzgIItZQUs4mlx+4MgJSvMceL1eAHRdPbeGEWXHsUzNOxqpoNDtIRK2YiOmi1MW5FLKVinlDnN5ENgH1Jzqcc9E7mvrGbHu38Wv+GrfhhnojYVFJj9o7Egux4wBbmz7Ih/avoOzn32JP/a6AIjEh2gbej5jP2NgAcTChIfCGIZxWvt8pjClY3YhRB2wHng+y7Z3CSG2CSG2dXZ2TuVp5wwPdQ8AUCS7M9cPzZ+J7lhYJIkYBj9p7sIhBA5D53hRHQPecu4Z0GgMR/lCWwVBvHQEmijwppK+CQoID7oQWhiJJBKxtPLpYMoEuRAiD7gP+IiUcmD4dinlXVLKTVLKTWVlZVN12jlFSzjKJlcnn+c/uFBununuWFgkiRiSWpcTB5KY0DiaX8rNCz+e0eZd4pf47AYr3I7kupLiy4mH88EeBKTluTJNTIkgF0I4UEL811LKP07FMc9EooZEMwYppYvr+PNMd8fCIkm+TUNHsjAcoCrQB8D2gtUAfG/fHcl2/82deGzKXi4BuwPcnlKEZmC3Ry1BPk1MhdeKAH4C7JNS/u+pd+nMJSolxAcBqCUzeEJKa8LTYuboiMZpCscoaDzKe70iY1u91sVb5E8BOCoWE8POTq7mZnEfR/vyKataCsCGjX+lv2f/ae/7mcBUaOQXAm8BrhBC7DT/rp2C485ZIoZBXyw+Yn3MkGhSuXNpGNwpb8MnlWAfiOsj2ltYnC5aIzEA/JEh3nTOBr60JOXPkL/hWs7j6eTnZmo5UfouAP7eUMfXi87ij7wBl2uIzi5L15sOpsJr5SkppZBSrpFSrjP//j4VnZur3LTrCMuf2pOxTkrJ4aEINhnlHm4EoI5j3Gj8EoABy3XLYgY5PqhC7dfMq8bv9+O325LbjMI4hfTx593/DkAjCyh2K7fExwpreTZscJ94EwB63DKtTAdWpMlppq9/O0/3BUas/2WL8lTZy2ripH4kPqke/J6Gg6engxYWWXh3QwsAVyxbDIBdKPPKZUX5NDUps0rh4704jBjNxiL6Dytf8t7C0uQxuihF2o4TN82HFlOHJchPMwcOfjW5HE8L9Nncqx7uCC7ybXb64/UA+IQywXTvf+k09tLCIkV6KH6NT6WlvaAwj8uK8vna0pSJJTboZUFLE237a+nauXvEcR7nKgAONPz3NPf4zMMS5KeZY4PtyeVQWnDEtn6VXEhDsqkgj+Pyy3z7hfficyrtvOvAi6e3oxYWJlt2q2fvFYd3UTpvAQDlLge/W7eIQkMFCQ1GfSz88wOsK+yjcXEt2qUX40SyIhbm5Zv/TKXRSitVgJWbfDqwBPlpREqJYYSSn0N6SpDHDaX16Ni4sKye0oJydneuoExTmvqHLr+BId2KirM4vUgp+cuWbQBcsHTxiO2P7PoHABHnK/H65rNyyXm0uUpoDHSwxu/jsZefxxsKXLiGwgTIA0B5K1tMJZYgP41Eop3Y0jy30gW5jKvwfAONUF8JlWbJt3wjZU880WaVy7I4vRw/fpwTcWUCvHDNmhHbdxzZDsB1574HgDUVtQDsdVSQH1aa96brXoetK0ojSpsfCoyIF7Q4RSxBPs3Eojr7nmkl0BuheeAYelp1vb5YlJaG/XQ3NWKTKiHRIg4SjeYlBbme5j5+5Hmr3JtFJlJKTgxNvUdTUNepfGwnrz3QyiMrzwZgUb4vo81Tex+m0NlK1MjH51NCusaVynDoi6lJ/aoly/B7PPSLInawkVBkGzt2vJlDh78+5f0+U5mzgny2BNCE+qM8+ot97H+2hcO9+zHSPFKOhYZ45Kf/j8d/+WOksLFAHuFV4SdYt24dFQUqCVHckHzN9ycAWg81zMQlWMxi7m7p5pzn9rF7MDR+40nQFVWT7G1ONbm5zGXHY0uJCyklobb3sa5sL2XFqaRuhY7U851fuTy5fNMCZR//hvgPAHr7nuP48R9MaZ/PZOakIG9pvZdHH1s8K9ycCso81CwrZN8zrRztO5jhWtgaDtPT0kRxdS0xacNFmJfVnocQglKfC7smiOuCRa5+ADr6QmBlj7NI4/aGJgA6oyMDzE6F2DBFaH2RKjdoGFGklDSceASbUM9ieemFyXbp/uXFjtTo8zUb1iWXn+CyKe2rxRwV5M1Nvwagf2DXDPdEseKCaiKxgww2NmZo5P3BQeKRCHFvPjo2XETI/0cZA/86jqYJyvNdxAxJnlC5K771slv4zM4XZuoyLGYxLk2M32gShMxI4nkdTdQIg0/UV6LrIR57fAXHjn2Pp178bbKt3782uWwTqX5cVpSfXC5wOvhY8wEAnuDyKe2rxRwV5EJTmoA0YjPcE8XCdaXUv+ILrC96Ej1NkPcN9COFYEvrCeI48BHG3leIe2UJABUFbuI6CFLa0U/6bSOOb2ERmeLiI1HzcFe2PMHjG3xUOnQe36ySZB05+k3qvI8DUFP9ZgoLN2Xs+89NS7mpqpjzC/My1l+2tI5VcidRrEpBU82cFOSGoSZ/pJza4ebJ0tbxq+SynqGRB4iWVjHg7yaOnbyIjbwLanBWqx9Apd9NVJdIJB/MswS4RSbdaeaUyBSb3Dq7VM2A2rX72LLtWrp7nsjazutYNGLdmnwv/7t8PvZho4T8fD++SJhOypkdM1hzhzkqyJUHiDFLBHnDwVSaz3RB3hsMEi2pQlTnExcOysNe/C9LFZGo8LuJ6YA0+I/1K1jQ1szSzqPQdwILi9+0pgqQTLVGvn+H8h13on5LL774vqzt7B7PhI/p8XhwNTkZEIV0Uzr+DhYTZk4L8tmikaeTIcjjMTSgz65uQ3HFQjRXaoKossCNbkDcMBB2O6883kBDWT0HHOWnu9sWs5Bad8pEMXSSGvmRUITjw9wX49Eo+/fvA8BBNNtuSbzehRM+l8fjIT+k5nu6KZlkTy3GYk4K8mhUaSrSmHlBHov1Z3xOF+S606C0uJQAaghaVp1Z0q3S78ZAEDZT3r7RKbDH49zTbJXKswA9zbPkZDXyC57fx1VbD2Ss2//Mkzy98hzKe7qoZOwgtOH28bHweDx4owlBXobHNfGXgMXYzElBrusqEEHKmc/hfejQVzI+JwS5W4boc+Sj563BhhLUha7MYWqF3w0IonG1vXrlCsp7u2nqGlmk2eLMIz1YLHIK6RsGzX2Dj3+D1z71LD9+ditdxRW80xbFSabDgN3uP+nzeDweCkIBimQ3v+TtNDtHRopanBxzTpCnC+94fGDGfclj8ewa+VIO0OaqZK/bwIka2hbY7RltKwvcSCmImK5gnrVr8AcG6ey3QpwtMjXyhN+3EQye3MFiYV548SGei3n43XlXAzC0aKR3yfp1v+Lcc/5Ofv4qVp31nUmdwuFw4NYEt/EVQnj5a3Rjcls8PoiuT21Q05nEnBPkRpo55eChO9n8xLqZ6wxgs2Vq2Qk/8sWyAUPYuFzuZRmq/FW+PfN2VPrdSEhq5D1DQaKaTsdQePo7bjHr0ckU5DIa5cDZ59D1gx9O+ljR48/y++IrMtb1PzMyZsHvP4u8vGWcc/afqah45aTP4/F48G+5hDI6iMiUmXHzE+t45tkrJ308C8WcE+RSzg7f8QSG6cv+h16V8S0R2bksfhiAI/UlBIWawa90ZWaF8zhtaMJG1NTIi6prKBro5kReAVErwvOMJ55mWglHo0SbmsEwsFdWTPpY247s5A/VV2esW1X1ZHK5qur1XHLxtpPuawKP10swVIKGE0mme2I02nHKxz9TsQT5NGMYEQKiiN3RYi684CkW1N8GQJkRwGHEiFZcQretDoHM8EJI4LClBLnT7aEuFCDqcNJm2cnPeNJNK4ODg0RPHAfAuWDBpPd/al/KlbFURFgvt1KRtze5rrBgIw5H0al2GY/HQ9QRBymT44nnDluT96eKffwmuYUxCzxV0jGMCCE9xvLiNbjdVThcTqARhy4oMvrY36cTj7sopAeXNvK96rBpxNIqtNSWKu29dc9e5l9x6em6DItZSLogDgYCxI5PTpDH0jxdntGUB8mX5Me4sPYqmhp/kty2ccPvKSjYOGL/k8Hj8RDQ+hBAQpTHYn1TcuwzGUsjn2Z0PUIgFmZZ0TIAhpqVR40Ri1AfOcbmQRcvspZy2rPu77DbMwT5whUrADi4Z2QpLYszi4TXiqbrBEJBosdPoOXnYyuamOYcT3sR7Fm0jDoxwAKO09SSaRsvLNyEEFOTy8Xj8RAhipAyaVop9Y2sYWsxOeacIE8EA80WwrEBoobBihIlgGWRSk8bi4W4LPA0BoI+UczCvOKs+zttNuK6Ttx0EVu4TKUGPd6eXfBbnDkkNHKHEScUChE9fhzn/PkTFrrpGQ4DHh/zPKbCoO+Y8r4m8Hg8hPWoMq2Y5y90pTzLZkv66Vxjzgny6YrmNIzoqC+Jjs6HaGm5J+u2oXiAGCKpkRtuNdkp4pI60YpdKgG9rKh+xL66odMR24/m6KUroM5d6vUC0KbHrYf+DCchyJ1SEgqHiZ44MWGzCkDvvzILlby5oHFEm+Lii0+tk8PweDzEDV2ZVsz5eo2Uy6SUs0sRyxXmnCA3pinj4RNPbuKppy9KfpZS0tn5T6TUefHF97Fv/+1Z94vGg+hSY2GhskHGTbukiBv4CwQLB1U5rHX57hH72jQbIBAiTtuAcjnMN/M993p9hA4fnrLrs8g9EqYVl4BQOEysuRln3cQFOdU1AHz1wP9wuPggC4OpKM6orZTLLn2R9evunsIeK0GukJDwfTdSKQJ0feqrHZ0JzDlBPl02cl0PEoulZvY7Oh9k94vvpbHx7jH6YoAewOPIx6GZ7oeJ4bDmwJ0H1zXv4nr5Oy4rzs96DKfNhdDitPUrQe43fc33LFtPw2NW6bczmcSz5LbZiBmSqADH/Pnj7JXCtnSJ2l9G8S26jLaj3UhD4+vNfuavvAubzTvlffakJdlKjCgNPRUXkS7ULSbOnBPkxmma7IxFewE4eOhLo7YJBo/gFjGEJ1V9PPnjc3oQtigrvIP8G3/ArrmyHsNl92DX4rSbGrnT9GxpK6/lmsXnWuaVMxgDsAtw2m3oNhtBl2NSppWkUuGvhrwyFqwpxBAaQ04/K0pXT0uf0wV5wmkmXXgbhhXsdjLMOUHe2nIvAF2yEACbzTdG6/GJx4NKsx6GECO/utbW++gf2EU0qswlbX3Ks6S0IFVBJSnI3XkYRoQlZysPA03Lnmzf5qyk3BGjZZSw/F+3Wv7kZypxKbEJQb7TSczhnLQg7w+qZyqvQk2gNzX9BJuIc0H1BWhZnu+pICHIhZBJG3m6IO8P9U7Leec6c0aQS0Oy45/HaWlVk46t/uupKL8OIRzj7Dk6uh5h8xNr+PvWd4zYFouPFKwv7fsk27Zdz/NbVOhyY9+LANSXpHxwg6EQwjDweAuIRNowtDYAhMju0u/zzsOlQXP/0eS6ek9K6H/tSMtJXJnFXECXEg1BqdfNkMvLUH7ehF0PAdqb1fNZWbsaIy1S+ILqC6a8rwmymlbSBHljy8PTdu65zJwR5EOBGDseOo4eUW58CwpX4nSWAiefATGRRdEVSFVHkVISjwcIh5tG3S+hkXcGlPB9Yl8RhjmODA4Mohk6BQVnAdDaqkYQo7mMFbjL1LFCqWISf1y/mM+3NQDQEdMZiM98lkeL048uJXYBJU4HEbeXUEH+pPy9W9sOAVA1fz3t/an5n9MnyNV/3YiA6VM+ELYSZ50Mc0aQe/1OrnjLCuJhFyc657GwcCFC2E4pla2uDwFkZISIRrs40fhTmpt/N+a+gcAB7MGdhA3BDx7vZF/bgLl+EE0aVM+/esz9ExR5VBGJwVjKo6DK5eQt527k6sf+CEB3dHYFQVmcHnSpih0X2u2EXW5CrsmNPht727FLnRKPh4OtSlEY1B2UeKav6IPT6UTTtGQdWiOiYxgRNM1DMOYhNNQ/zhEssjFnBDnAwnVlaJ4wBYFa3K0lILSTFuRH928mPDQyB8TOXW+jqvK1uFxKwIpRshw8v+VaCsQgId3Br955LmdVFwAQCAawGQYFJSNrHWajyKMSIEX09oyJTW/9Qtx2ZWLp6Oub8HVZzB1000bulAYRp5vvvuptxCeaTK3nCF048RpRNCE42HEMgDbvy6evw6iRp9LKVYC+3hdGGjE0zclQ3EM4agnyk2FOCXIA6QwR1gye+dUxVADZ5LMEdrU2cqTl7ex44Y0jtgUC+/B45uN215DnOAtbOLvbYAKXs5Bz6lNRm6FQEJsQGelt6xZkr4cI4DBdwDR7D4ORVLCTEIJS01zT2DJ2FReLuYnSyMEWUCbA1vJavrL/2MT2PfgIje5KfA6lDBzoVtGc6yrPnpa+ppMwr0gE8d4IUsbRNDsxw0s0NrP1A3KVKRHkQoifCiE6hBB7puJ4p9QXGWOgvJtoKE7rC90npZF7C0zhL0Z/CUSD3YgmL97I8jGPlefJrK8ZCEdwi0yb+IIF7xp1f82mAoWczn7a+zNdsxauUnb2I0eOjNkHi7mJjsQuBLb+vuS6w/0TE4S/aGrjucJ14FDP19l5fwZgWdn6qe7mCDI08t4whowjhB1Nc1Lp2mq51J4EU6WR3w1MzOg7jUgpsWHg9edx8WU1FIV0wBjxYBw79gP27P3IqMfR9dGrrHi9ixja30M01I3DUYSzbuwq4n5PdXI5EgoyZBh4hmU5tNtH1+ptmvqhuRwDyejOBAsvVtkP27q7R+xnMfeJS4kmQOtLuewFhoYmsGOUe+3KtPe68kLiukGxQ+3ndZVNS1/TSQhyBMR7w0gZRwgbFR6lkHR0/mPa+zDXmBJBLqV8Aphxh+a2YAuagEJ3KcuvX0RcqHD2zsa+jHaHj3yd9va/jnqceHz0bGwuvYquX+5BdwTJX70Ip2fsivb57srkck9LEzG7A69j4tmDEyYYtz1EU2+mtuVzKyE/EIlZWswZiCFRGnnai3xCE99NW+ix+5kf6+Zzi6p55sQ+uuKCAVmByzX5ohSTxeNxoxFHCtB7I0ipZ7jfRiOWYjJZTpuNXAjxLiHENiHEts7O6Ukkf6RXuVMVe0rRNA33YhUtuf2XTxGLTtzEEtezD08FNsJtXdiqBQgDp7eExYs+xY7em0c9ls2eCkjqaW4ibneS58oe/JMNu70QAK9NcqQ3M6mRU1PmmT01C3nfCweG72oxx4lLiQ2B1tWVXDc4gSLMPzl8hKPeWor8ZQSCB3jkyIN4hMSfP/32cQCPZwBNi6FrEY67v0k81pchyAMRK7pzspw2QS6lvEtKuUlKuams7OSHb7HYAMeO/zCr7ftYv0oiVWJqyXqRemH4yx7mmXsPTfgc+igauT1YinRH8d9YBYDDXojD4WfD0lePeqyEaQSgp7mRmN2J39SkV531bdat/emYfXE4CgG4sThKKPRUxjanaWc/VlrF/f3hZEIuizMD3TStiI5UibTBcX7SLwyE+Iymqte/vKKcLVteyUb5Y1wa1BRWj7nvVOHx1KPHHQjs9PmfoLvnCSKRVFrmlhNfPi39mEvknNfKnr0f4vDhrzE4uHfEtqMDSpD7HMrmvGKFeiCcnn5OPNXMkZ2ZI4Fs5oju7s0caPiv5Oe4SNnAxVAp0h8nbCifW6dTeaNcuiI14ZmXtzLjeL68JcnlnpYmpMeDx6ZMPhUV11FSMnaVn/TERQscj2b02WPLvH2P9GQP47eYm+gS7IF2tM6UIA/ZncT10VM57+hKtX1VmXKJ9Wkx7ALczrE9sKYKj8cDUgAOFj75P3jcC9D1zEAgy1Q4OXJOkAcC+0fddsIMYxeaGqY5ncUUFVzAQM1TLK/p42jPeRw58r1k+2yZEnfuejvxeMq04q9+Bz7PUgB6QyVEYifo7HwcgMJCNRS1m+YTt3seXk8q14XQPJSVpvxye5qbiLk8+GwT/9rTvVsWeY/R1vbH5OcyZ6at/UDQGpKeSehSIoJ92NIqSEWdLv5+rHnUfTo71ITiOUNHWOjOFJbTke0wGymvFYE9WoDbPo/8/BUZbXbtfudp6ctcYarcD38LPAssE0I0CSFGJieZIhLFHbIVeWgaPG72JyXgevufAWBo7dcBOHrsf5PbdH18wbe0/CI2bvoNmzbeR/FCNQnT0vprhHBkPPhnb7qfszfdy/LlX0yuKy97eVIQ6/E4fe2thO1O/Ezetz1BX1+qkrnP1OwTfOmI5U9+JqFLiS0apWhABdGsdYAzGuah9tH9Dnr6WimK9VNauwZ92FyQpo3MiT8dpPzIw8RdvRjRKJpwsiAtnqKo8LzT0pe5wlR5rdwopaySUjqklLVSyp+Mv9fJkRTkw7TpnnAPg9E+ALQsCagM20ih3dh0N729zw9bm9KANxtnU1y4CYejiIKCdVTWXJJqJjPP4fevweksxeEopMPwA2BLS03b196KrhsMaRrx3/wGIzrxSii1tW9J615mutsvL62d8HFykaYPfpC2L3xhprsxK9HjEUTcYNWRBm4s9vHfdRXUtJ3g6bCe3TQhJR2hIE4jxjmF/oyRJ0Bx8YWnpd8ejwebLY5E0LrqLvqjWzCMKIsX3ca+3g2ntS9zhZwzrSRKQUkjhmFEk5GbR/qOkNBPhbCN3DGLx9/Ro99mxws3ZTazFySXF1ddm9SoY4Zku/dWhF1VsddjdmSWyUUpJcG48sl1uVOTRz0tTUQdTgyhUVxWhuacuOfKksWfTS6H45m37G01pTxbV8CG4wdUQds5ZluMNBxE77PCtrOhxyNocQOH28n/rlnMuTWVnHV8P23Czp7ASH9y2baHp/LPwmVEOafQR/vgseS2kurb8XjmnZZ+ezwehGlaieapUeRgQM15HR96GcAIm7nF2OSgIFf2QMOIsvmJtezd+1EAmtr+SqHdLKOWlro2YWbRbRObCHTY/cnlc6vOTS4/3N3Ph/c38WHjq0jA7fMgtOyZ5pYXqBJaPm+qDmdPcxOdJcqnfP6KpRPqSwJNS72FBrMUUJk3bx7OeAwpBKGJ5trIEfSBAbQC//gNz0DisQgiJnHWViOEQAjBctRI9Y07R5YBDHrLCNh99DsLWZ3n5cWOLclt1SWLR7SfLhKCHMCwZ46Ui3zKXDldtXfnKjknyBNEo50YRpT2jgeIRnvI7/kVby9V2nq6Rl5aeuWkjiu01EtgSVHK4+TYkDp2pyyklWqQ2b86IQROM1LTnabh9DQ38uAVrwfg7E0nHwbdlSVyz263k292pyswelRqriGlRB8YwOYvGL/xGYihRxFxibOuLrlukT8PgN4sqY27nCpXeXl+EQ5NUJen3HT/1vh+yssum/b+JnC5XGjCQCJGmDxLfMp+HtOtjJ6TIWcFeU/vM8nl/n41AejTTI08TYNdueLrEz7mwMBuQqGUJpNeJaU9nFKFO6gg2Ds4qhnjrJX/Q1XVG8hPc0Vsam+nz1/EaxoPsaR48oJJlL0XgL5wV9bt1XnKc+a+fQcnfezZihEMgq5jK7AEeTbi8RhazMCxMDXCK6us5vwdmwEwhj2f+0xzy9p8JSy9hpq89+Zt5HSiaRrCFOTDKclXz3FPwDKtTIacFeQdHX9PLu9+8b0Z29K9VrK6VMksNnTgROPPkss/OvEmjnSqwKBHH9vMD5tTYcMhfLjy1VC2LRLjN62ZIcU+32JWrvgKmqndSynZE1M/qtfXnlwI9Ir6m4kaEDQndIdz3Xxljz/aMncqBiVs4zbLtJKVeCSCZhgZ5d2Kqqqxx9ToMTZMkP+rW5kXryktBKCp5QHaQ6XUV0y8YPNUkTCtJFi94gcAlFqC/KTIQUE+fgWUdNNK9oop2cP1E3lNgjq8eOIs8lzqhfCvY5lufUF8GDLAvsAQ657Zy8f2N9IdHd2mF+ztocujzC2rzzk57afCV05MQiRLiTmAxQvmYzN0euaQidwYSAhySyPPRjymY9P1EYJcmPNI8WGC/FBIjSovKVLml1A0TG+4kOWVpycQKJ3EZGcCP8pbpdyvFK+ekCXIJ0POC/JELpJ0tGF1OmtqbhrRJuuRzf0kGh+57DzK/cqvturEsYx2PZSwW67l8q2p/CYdYyQr6jp+lKA3H00alPnGzpY4GprQiEsbhsyeB6asrAxXPEakqPSkjj8b0QfUS0vzWxp5NuJxw9TIUxp1YWU1NnPCWx9m+TsxFEUA+WbStlg8TjjuZtkMCHIgJcgNDaNPLZbnmzb+4ASyOFokySlB3jEYHhG0kE3jHu5+WFh4TsZnJxWQxbytG+rrGIh5ufXCOsAMFW5WkXLOWIT6wW7+Ka7nm8btGftevvUALwxk1yJannic1vJaSlAVXU4WHTuQ/QG32+1Ux8I8pXn4feuMJ6KcElKmlcKZ7chsxDCIG2ZhieJU4RJPvh+n3RTUae6xEcOgPRrDZXpaJVx3dZyU5mXGJpwODL0aaYofzXBi9KnRgtel+tIfsgT5ZMgpQf7dh7dgGJnCMh4PEheZwl3TMh9MTWT6bOcXrsxQ7BNVvJ9sVDbxQMG7cZhh9PGWFiJmCPQt936Pt9bXEkZw8zxl6/ZFU7PuDaOEyP+zJ8DxeYvp1bLb5ieKxIlNC486yXqzUOf/4uG5YSfXBywb+agEO4gKBy6HlqHMCCHIy1ffl572nOweHMKAZHqIcLgZpy2K0356wvKHo4k8IlFlDxeGk3iv+g0mRsX9ISvdxGTIKUH+ygX3jFgnZRS7zLzpwwV5WdnLWLL4P/C41RDU669HkBKq4XALzX1D7G1Xlelfvy6VzTB8oIGAqSWsWbeRVbVqUvGnptY75HSzzIw27Y+PtJMbwSBH7Wr/D84/tVzPQrhwaHHaBrPbyW2lZVT2dRGI6xk/4lxF7zcFuWVaGclAM1HNgXtYSuTNmzdjuJVwTtjI9wWGeNxMqDYQV2aXru7ncNujCOey09jpFDabDWmoF5AmXei96jeciJkYGLIE+WTIKUHutk/s5g4X5EJozJ//jrRkWqUZecL3H/gcd/7tJWxO5dpX4kml2Y0c2M/RSuVve/H1b+DcgryMY391aS0HNPVj6omNnEQdfOxxBjxe7IbOJ+srR2yfDF5XPg4BXeG2rNv3u/Jp9xczJCUnhiaeAmC2YgwMIBwOhOfk5hXmMrL7BFHNgcuX0qgNw6CtrY1es7brUDRC1DC4fOsBftasnu2EJ8uJDuWmWlk8ueC0qcKmaUnrpiacxPsSGrlSsAIRy7QyGXJKkCdMIOMxXJAniMWUhud0lGRMiPb2Psvf9zTidKqSWemRoQ+3d/HQxdcAUFZehUMTrPApU85nF1bxlppSvlwSRUiDR7pHasoNjzzCsxsvpVjqo3jQTJx8dz52Ab3R9qzbXzuvAmmWkeuN5X5knN7Xj1ZQcMrf21wkdmQ/MYcdt78wuU7TNF7/+tdTak4YPvvM02zrV6bIhJLx6nLVvq1XpWJeXL3m9HU6DZvNlpzs1DR3UiNP/PYisSihMTzBLDLJMUE+MS1zdEGu/L19vkXJosZSwonBxVRXtCGEDmiItECgv3lTM/p2c6LojiU1vLKsgPfMU5r6rS98kXmRdqLtL/GzYyeS7fWBAW695DoAbsibeHm30fA4/DiEpGmwKev284v8OM1c1IEJVIqZ7egDA3Pa9XD3YIido0yQj0f06CFidgeewuKM9TabjY1rVgGwdfcert+ZWVBlfb7S4O36HjpCJaysXXRS5z9VlEaufk92r5f8S2uRUiZjQGyaQUufZV6ZKDkmyE/OtDIcn29p0pMlEPNhE0HOWdmFQ2gZIfote1/kWKmya9+zNvXAX1SUz09W1ScFu7jmaxTZoDTayzn3vIanH/sR6HGOPPI43eYP7TPnbZj4hY6Cx1mMVxO0BLJPZto1wTIzK2RfLDdDnHtamtn21z8SDQ+hD/TPafv4y7c1cPX2hpPaN9rYTMxmx1M48kVXWFwCQFH+SLdCj01jcHAf+Y5Oyr3deJynNgF/sthtNqSZ5kK4BPkX1yKEMH+7Nnz2IC19lnllouSUIK+uesOE2o0nyG02L8uX3Undkh+yu2sVha5+WiO7qfCUJk0uT/cM8o/f/Jy2shqui4e4pHgMX9vy5RSUL+Hpog1cefbP+DfO5v5fvo9vNSsTyE/PqptQv8fD6SwhzwbNgdELB1xkupJt7ewdtc1spuvEUTb/6qf0tbWi9/fPaY08QVskRmtkcnMaQ209GDYbLu/I+QOPS402K1x2SuPquDbTIu21aURjM1/c2JYmyNNTUmuaHae7nuq8Nlr7LUE+UXJKkPsL1o26rabmzcllTXOM2g6Ui5bN5uL/nillMJqP1x5io7+IJYX1aJqD3licf9t1mP+46mYCeX42VY4fZHNJUaagf++C93HfWaqC0KXhidcLHQunoxiXMAgFRz/eq6rVRO1zvdkDh2Y7BeVqQri/vQ2jf+CMcD1c98xe1j/z0qRSEAdNdz2nNvInnBgpDvR0YZgTxbppxvBoGqEh9ZLfO/DGU+r3qaA0ctUnn29Jxjavu5x8Z5Bmy7QyYXJKkKe7DA7H4Sjiskv3cP55j2TYuEdjy9Ee7n+hmTXz5iEEnMNzFMoOnM5SmocyJ1U31laNe7wPLKjg0bOXcbbfh33Y3Jzvp1dC665xjzEeiQfeFh9dI19bW41N12mIGTmZmzwpyDva0Pv70eZw5sMie+bznM3rKSuGTtB8T7uypFJOPH+9fX0MDMuCeLzhAMc7Vbm3sqLTl7p2OJqmYUgbmvg4K5bfmbEtGNzP4sKjDA6OXtbRIpPcEuRjCGhN2LHZPHi9daO2WVj/ESoqXgXAj548Qk2hh4uXqYdZ1wMEgwfxeuo40pmZYXCV3zfiWNlYmefhrxuX8NEFKTfDs/Pd8JrvQ9XaCR1jLAoKlJ39LMcggWggaxubzUaRHiUiNBpCE/PymU248/Jw+Xz0trXS5PbOadPKYq+bEkdqErx9jDQP6ci+Fjpsau6lwD5yEt1hevk8tXA18WHv8n1bnqelcz+6obG8un7EvqcLTQiEphEO141IbBeLqRgNW3z3THQtJ8kpQe7xLMDpLKOs7BUjtglt/Io79fUfZNVZ3wLguzeu52dvOxuvpyijjddbz9/CmU//8Gr141Ga5gP7+/VLYf2bx2g9cXpb1Q90uccY1XMFYKGmPFb+0dk3Jec93RSUV/J4FN58x7d5vLxmprszbUgkq/I8/Hm9UibGyteTTqxhF0erVXDbiryRdTY9prnlUL1Ko+xL09pvee2rMeJNtAYrWFK9ZMS+pwshQGgaQ1ny6ycw4tndbC1GklOC3GZzc/FFz1FaeoW5JvWAOh1F2XcaBbfDxtKKfBz2zP083jr+3NEHwCd/8QP+r290gTkaVz//BP/z7Ts5NN+Pd5IvgbEoKk/Zi1s6H0uWuRvOUjMxV9vuBwg0PDJl5z9dFJRX0BpT1/aT4qqcNBFNBAP1BPtNE0t/lmIQ2Yge3EvYqSa1/faR5kZX2jPnx+CcwlQQ2+oF84nFg8QMB17P+CbD6UIAQhNZBfmyZXcAcGHl3+fsvZ9qckqQJ0j4msq0zFcOR/FozcfE4cgcugtbYXL5mmc3c+2CyWuEob//nQs0nbxFC0+qT6Ph9KSG0bLlfzh+/AdZ2y0rVi8nT+9BFjeXsLU/t6oGFZRXMmC6T+60ufhxU/ZiGrmOlEoz1Z56CoBAfGK+/9GjR4g61IS+M0uwlDtNAw9KOLfAx7FL1vD4Ocvw2jQEETShZZQQPN0IBEJk18jLy65OLveGctON9nSTk4I8UZg1/RF2ezKrye8LDE3obe50liSXvd6FaP4LAPhMm5oQci2dXAizjEZxVFZR+NrXTmq/ieIIfCq5fPjIN7K2WVGmXmp/Lr8cgL+3dU5LX6aLwopKgu6U3fSxnonVW801DCRCSgY/+x8AIyYmRyPa2ETczJWfzWvFlbbuhrYG3lFbhtumsdznoSsQwWML4XLMnBCHhEaeXZDH0iqlN/daecknQk4K8lh0pB9sft7y5PKTPYNcvvUAv20bP52rw5ES5GtW/4AIytbuaGnBMX8+tryJTXQmEE4ntd/+FsW33DKp/SZKcdX4I48Fpmml2a0mXaMDuZUNsaCsgoA3n9JedZ9X5M3NXCu6HkEP9+EJBhFSTlyQt3Vj5KvnNJvXijtduHd3kZ9mftnX+BKF7gHy3DOTgzxBxDBGFeTf2vFddg+pEUdrX/ZJfYtMclOQx/oyPtvtmQ9lmzlplC33yXDS83g4nSWEzNB2+4njuJfNTEKhsSitGV+Qlzszta2W/tzKT15QUUnQm09FdwdFNo3ABAVcrhEIHmIgtBNNShYIyc7BiWmf0c4guhlqn12Qp9bFujsytjV1qPq2lcUrmSkagmEe6xlkyDStGEamSenahddy3HS46u95YAZ6mHvkpCAvKVUmg3PPeZD589/Jhg2/y9ieyKc/OEkBYLcXpAR5UxOuZcvH2eP04/KMnz/aNWy43RLJLTujv6ycgC+fvOAAeXYbwTmQNyYbEhCoa1vhddEUHj+6U8ZixPrjGAVKeXFktZGn3f/BfobS0h539x8FoLpsZtLXAizxusi3aQwIjTgQjWZe97qydfjMuavi+NfR9dEDg8LhFg40fIFA4ACPb17LS/s+PZ1dn7XkpiAvvogrrzhMXt5Sliy+PcOsAtBn5gUPjSIAwrrBR/efYK9ZVXzB/Hfh8y1BCJHcxx0J45qFGvnwCklSjv+y6tR8EMgNO7k0JD2DOgP5RbijQ+TZbQT0uamRSzQE0LBkCZFQgP64jqHrPH3Prwn0ZA+jjzU2ggS9oBCXJrJmhrSnaeTOaITe1pRprS84gJTg886cD7kQgmV5HnQEB8vnjTCvCCFYUbo6+TkSGdsNsanp5/T2bUHXA7S2/mFa+jzbyUlBPh6JSj1RI/tk589buvhtaw9Xbj3AvW09LF78Kc479x8ASe3PHY3gXj77NPLheWQSE79j0eEsZmvD8zlRbOLAljY+9ePtAIRdLlyaRmSU+3iyRA2D/zzYzL7e/hl1b5MIBAa9NdUMtLbQH9N56cnHeO6+39J+NHsahmiDihCOFxZn9VgZzoZzzsPlU/M8uiEJRoaIG3Y8ngXj7Dm9VDrtOIAdC5YROfQ43HUZBFPeSavK1iWXI9GO4bsncbmqcDiKGRh4cdr6mgvMSUF+3CyqMGSM1MiPD0X4r0MpDeUD+05kbA+Z+3g1gaNm9gWj2O2ZhS1GE+SfWah8hDUgrtl5VX8tP26a/Vp5m1dwsEY9llfGH8SpiYzak1PB3zv7uaupk5dta+Cjjzw5pceeDEqQQ50h0cJDRKTkqT/dS3n9IhZuOCf7PoOduApihAtLJhSodvUNN1NSMw+AEz0h7CKKROBynVq1qlNFCEGhTTDg8fF4EPb193POtiN0mvNbFb6Uj7uhjx40JITAn7+KwUFLkM85EiWu0m2rcUOyYyBIZ5Zk9UaaVpYwrRTW1iCyuHbNNMP95UcT5AldrSjNzawtB2zlb+pu42CV8lKZv/goDmSyqs1UsX1A+dXH7Q7+YMvjQGBmsuwlNPK6ggI8hjIftff3ccEbbhq1mEb+ynIWXtNJm6uAatfo0cwrzeInBWkeK7ohmefvR9PEjPqQg3o+PTaN0sE+vsYCvrXwnZwwnFy9TaX1TS/u0mja9QHi8QB9fduSnw0jRlwPEUxLJGeEz7yCFLNPUk0BCUGeblv97vE2XrX9IJ87qBJO2dK2pQv3oDlBWlA3s0PP0RielyI6Ss6VRGGJee7Ujz0U6pu2fk0HefkhRCw2qonsZHm8Z5A8M72rjsZ/HBw9Cdl0oiY7wVNTwfwS9YL2LVo2qjYOwIDqa4vhoMY9epbPe9Yt5t51izLs5YvL81hSdBiXfezsoKcDgcqdtOn4flqkxv58lS6gOYuycfT4jwkGVVzHnj0fZPuOG4jF1ARuR8eD9PdvA1JKW/v/bkdOMLhqrjAnBXlCgxuIpzIALnrkT9jjMV4wXbwubD+WbL85LeVroK8PgMLFM1M5ZTyGa2rdLdnNJQm3tHMLU37wP++O8eIEXdxmAwIgGplyjTyoG9SfaOBl8kEAtvYH2bt/P5+//6882nb6cnWryU4DW205pVIpE6XnXTp2abuBFrB7GNAlhVnC8xOUOu1cNCy1smFEMYwwbvfMmwyFECAEC7rbqEfnqC2VKqM9EiMeT/0m/bKF3S++G4D+AZVIS5rfVzg8MoWGZ20Zwj4nRduozMmr1dN+96/Y1oBh6DRveZpImiZSs3R50t+2Jc3ta7CrG5seJ3/ZzLlnjYfdnkor0NmYXZC/d14531w+j7fXZOZS/+rR7IWbsxGN9jA01HhynTxFRMIbJxKeco08FIuhDQW5lR/zRvkrolLy6Wd28IPCedy07/RerwD0cheBXVsAaDfGmcDsb4KCGgZ1g7wxBHk2hoaU0PN45p9MV6cUpZGD0+HgNbEBYiJ1LbsGQ8TifRntwzFlDksI8JQGnvl9OYbKyb80M8r7TGBKBLkQ4mohxAEhxCEhxIw7csal5HKzos/uwBAtB/YR6u9DS9PsfhuIcUu1EnLfOt6e1NwHe/twRyK4Fs9crubxuPCCzRxwvxKA7tbseUjcNo0bq0ooc2YOo//VPcBTEyw68cyzl/HMs5edUl9PlvXsAECGh4iNkhzsZIhGowzF4uBVCcguRyUV27Pw9AfIGKZGvid+O0arMpkcamkd25NmoAXdX0tQN8i3TU6QDw7uBcDrqTvZLk8pUoLH42FZqB8hNF7W8zwAt7x4lB5fZobTYDwxklTPgiHjBAINRKKZiozmtmPLGz8T6lzjlAW5EMIGfA+4BlgJ3CiEmLmwMZQg99tt3FhVTJnTzqGtz6I5nXhNu+i6PDfvrC3lLtOLI2xIjpmeLj0D/fjDIWw/OhcGJ669nk7s9nz8flVgNzB4HGOMgBmvTWPLeSv4SvyF5LrX7zw8ofPo+ulPtuUQghVD/XwQlUemr6+VnsjUVDt68skn+cY3vkFU0yguVC9AP8rWOpQmO0dz0wwHUvMRA3Gd+BSMFISZ+K22Snlp9EZjtLWN8dwNNBMoqAMgP4v54Onmp+kJZ4/kDQbVRKLPN/PxEQI4Ho7SW1jK0NAQEuixpcyAf+7zcPllLyU/h+NBono0qZFLI87zW66hqennGceNiBZ6e58/HZcwq5gKjfwc4JCU8oiUMgr8DnjNFBz3pIlLiV0ISh12+mI6DVuexbVsDbqUvMJr5w/rl3DH4ho+UZcqANHzwu/hh5fQFY1QMtSjCkFMkQCZDqr9iwlEnZSedR97dv3PmG3ne1ycv/pyzmocu9Dv4OBLo6bGPRWk1EekVcjeTnmorIj34ERNesmhANEsbqQng8PhYN68eUihUVKYsq3+W3mmS2e2vODb//Znfvax9xDo7SFqGFzyzB5et+3UKtgojVwJ8itefT0Ae6vr2bVvX/Yd9DgMttLgV4K4ZpjXSn+kn/f+673cc+CerLsHQ8r7Iz9/1Sn1eyroiSmB/KP6NbRF40gg5J/P3XtUArHOaAwh0q5PSp7c/78YRtT8OLoH1pEj35y2fs9WpkKQ1wDphsUmc10GQoh3CSG2CSG2dXaemj/zfx1q5rvHR4/2ikuJTUChw05MSjqGorTHIepwsbS0mHy7DaHHuC28lf/Xqt7oQ1t/Sixq0OMpoCSvGG78LZTOXOL98ajJq0HX1YPe2fdDBgdfGrP90ooF1ERGL8jc0fpPtmx9FW1t909pPwEOHvwSTzy5EX0Mf2CAqKkJO22pF6gzqoOYGi+L8847jyMlyn+6h1SytG8uKeLlJalc7986NvLZqlu7npdKavjRz37M3x99ijZdsjUYOeU8MAlBvnTDJgB6fX7u6k5NSN+2/wTXbjdfwIF2kAYvupVf+KaCzIRuuzp3IZFsKN+Q9VyRsDLfeL1Tm175ZEiPun7WUwjAZfMWclbgMGvDjXRE4xmTvn6bxGj/cfKzIUd3Mezr3zr1HZ7lnLbJTinlXVLKTVLKTWVlZSd9nP5YnB82dnLnkVaCo4Ruxw2QTc3I3/wKgL75S3H7C9CFoDDhV33v2+D3N7O4Zw8AH132Ke5Zdichj4eC/JnNDDcRqvOqiYvUw7xl66vGbC+E4EPdasZ/lXPkbW9/6WEAhvpaR2w71ejHltZ7AeUDPBaJwB+nSOUGcelxdDl1j+l9hnr5BUhp4YaM8bqKlNfEQ10jk60V1szjDy+/ka+ufznvsaWEfra4hIliyJRGLoRIBnG1So2EsvPr1h52DISUGcd0PfxKVE3mVTjtDA1GaTrQS0/PM+xsexa7sLO6bHWWs0Ek2oUQdmy2mXc/TI/W3ZOvXC9vrCphZ8WFFIbaGIxnfq82AVpa/QFpjPze033Px3vW5hpT8QtpBualfa41100LsTSZcnSUmpRxKZFtbXj2q6HvkC+fi665FkgreHvee+GmP5B/658AuGH5eq7pbSfsdJNfOPvrROY78ylxT67K+FlnrebCnVuJBUcKKke5CsLpblZFDh65O6XhjzWMHY9Q6Bi6rn5Uif+jkfhx27X+VL/0KFG0jKCtk6Wtr48OtxLgt/DT5HpDj7DMl8ph0xWLYUiJlJKP7j/B5p5BTrRnTipXmL+cvpPUyOPhIaQUSUEO8MEFFVzi9zDo9rJvmHmlLRqDgWaGNCcDUmOJ14UQgmf/dJgn/nIXL+x8C77eP7KyZCUee/a0v/H4IJo2O1ICR9NMeL1uNbKY53FRs+Y6/PEAXVme0XQMI/Xb9wwsobriBvz+9cl1iYndM4WpEORbgSVCiHqhjFpvAv4yBcfNSvoPejS3tLiU0NmJXlsNwH0bL6epUHmo1CYCZOougqUvp8LjwSbUBFe0oYEhl5t8/+zXyAH2xDJHNm1tfx6zveui67EbOtHwyBeAcKkX3IC2ldbje9j/XGrCLf1HM1F0PYxhRGlrTz0K42rk5r21p2nkTt1MtzAFGRC37t6N1GzcJr5PESkzU2/vs1RHUxOEcQn9DY+yZzDEb1t7uGHXYc7fp3STWwNdvPuxe3jF5j8B0B8/OY38xQf+gpH282tv/xsAxW4XuNzs27cvYyQ0ENehv5nPLv4wAG+uUqah2tUDVGy8CwBN72VDRXazCoBhDGG3Ty6//nQxPH9OucOOx6axfs0r8BphQjF136OOy7Lu39Oaul+FvotZvuJOigo3khBpwdDEJvTnCqcsyKWaRv4A8BCwD7hHSjltr8P0n3N0FC0tbhjI3h4O5ace2o/uV2b8C4syJ7Y8No0FbheHhyKEDuwn7HLhm6R/7kwx5MiMPt370sc4evT/MkKY09GKynHaDGJZtMj0AIzjzz6FSIsInKwgDzzbwubH1rNr97uJx1NCeTwvmIg5qWkXKVt65WJlr56KVLZ7Dh4EoEjL1Paam3+LO+ThGz2/5KbAZgB6/vQh/vbgyEmzD7zsEt5zy1uTHitN/ZOfENfjMbY8pF5wiTS2TU2/BNTzKJ1OWltbaetJCauBuA4DLRzwKfv2ayoKCYWO0thzU7JNWwzWl6e00nQMI4KU8YwYhJlkuCBf41UKlnD68PiK6bPnETEM/FVf4tEHvj5i/82/351c3r1ngNBglEHXWp7lPAwE8VjuBL5NBVNifJRS/l1KuVRKuUhKeedUHHM00l3DRkumFDcMok4nxjAbul2ALUvUXLHDRl8sTv/R4wD4JumfO1NsqNg0Yt2Ro99k94vvHXUfl8dJDA3CKfNFIHgwY5JTtvdQtzo1GZjwFJgowiOQWpSenidID9iYsEYuI0hd3QO/X+0TOkXPFT0e48V2ZXcukZnZ9MKRFp5uOJvKoj8xP08F5vxj3ae5x3VuskpRgmqXg6rFy7j6LJUZ846GyQcQ7d38CIEB9f0nvh2hKfuuR9OI2+zsq1zAf+89ktxHCfIm7DY7FxTmUeVyMjC4J+O4C5ySZb7s+erDYZUobrJFyqeLhCdSwqq9Im3eZlGx8iZ78GgDtUV5/DlLSoHVF6buy5G1dlrQeevRMv5P3MbDXA3M/rxCU0nORXZOSCOXEs9QkMKGncl1f92whGfPy+7eXuiw0xMKJwtR5E0gq9xsICHIRTQzkVYsNnpFIFdBEXFhR3/pX8l1oeCRjDah1T9h6brUN20Yk/tROJelj3pS92g8jTxhKrPLKEZMpet1oUYDo+WWnygntj5PS2k1lUMBisi0d6d/X1WaEvL39ZXQUlDGjf2NNLSmNPPEk/Gyf7sBgH6nm11bnptUXzx5fupKK5NJswA0TWmkHptGREo2L1vPn/SUAOs3TSvdzkKKHOolF4/1Zxy3wmGwf9ebs55zMHAAAJe7elJ9nS4SGnnCMFWSllf/7StWUxzr41/NjVQWuAlkSV7Xp/0muXy//Wresf847ebBfiHemSwVd6aQGxIrjXQbeTyLIJdSEhcaUX8e7qGU4Di7wJeRQCqdIoeNnnCE/nzljVDinNnMcBNFmGHNTtdC3L1LYALeHa6yauI2G5Fn/55cl0jEle63e3zo35LLk53sHAjtSC7Ho6l7MJ5GnhDkNhnBCKuXgQM1Ggifoka+/x9/I+jNoybPO+b1lNOFBrxUrgRe9cuvw3/Tz3lgsZc/VIRUXDmqcPDtVcpM8fNHHufXx1t5dAKlBQGWnHsB5+eVgpnGFtIEuaYRzjLS7I/rxAbaOW4vocT0vIrFlSDPz1877jkHBlQe86Ki8ybUx+lmrVmqzmea8Hzx1D2xlSzkrEgLx0NDuOw2PvGKVLqMgeK3ZBxnwfx38Y1V6zhoOj5o5gthS0jNc4UPHGDgoX9O34XMEnJOkKfnUck22ambP/iQ005p7TwuKMzjq0vHzr1QbLfTZ0BfninIZ7jC+EQpKjqPysrXsnTZHcTidhDjCzt3cTFxm43w7u3JdYahJj+XLfxJ1n3G8tkdTih0lJ273p783N/YSWxICbzxvFYSw207MeSQx1xW547ETt7NT0rJ4cMH6Siroaa4CMOIQ3xe1raakBmjvjxTG9w0bykXr7wgo+3V85S74NHSGm4/2Mx/H2qmfvOujNw9oxFrbsYQWlIjT7jODc8xXiTUM/5wZy+NMUlU2Higow9QGrmmeShc/JWMfbKNoCJh5Vbqds0Ojfyby+fxwIYlvLVCPRtF8cx5mAKXjz4diAR4/+WpdBkXL3prRjvdCHNve09y3HejV6WzjZlpgft+fw+t//EfM1pA5HSQc4LcSBuqZ8uKFzx2DIBwNMKScy/gj+sX89ZhiaOGU+iwEbLZuP/a1wK5I8g1zclZK79BefVSomJi+SWceXkEvT5+XXw2Rr/yxEgE6hzfnX1SU07CRv7sc1dlfI729WB3quOOq5EnvVaiGOFMQd7f3zfhPqSzcyDE9c/t4W/nXEnE6eL8Qh9gEGw9B9dzv0SLDXfHy5xDaTWjPKWU9PQ8nRH5Ot8c4T27eDVRm50DoQhDhuThCWjm0ZZmQOBx1yA0H1HTO8c9rJhyr3m6bV1d9DmUlvmGymIMKYnrQez2PHa07+CPh1Yk9+noeHDE+RJVdmz28Wu+ng6q3U42Ffj49KIabnnmQQojqQnurv/3/9B2N9Fny4OjTwCwZLGK+Cy0V9GdFsylx8Pc06Y8kISU/CFYj0NG2RVW35Vr2TKMYJCuE018+UjrlEUJzzZyTpBnauQjb0pgp5rNFkacxWefP6FjJoovPLNE2dBzxbSSjs2bKZCMYQETUkoGA/uT+anvfPn7+dcBZQJJuAge2pZdAMlJaOQj+uXuxelUZpL4uBq5aVoRMfSoMhslBXnv6FGpo7E3MMTV2xt4Nqzzwmr1LMwz3Syj/TrzAwb1nZ8dtpfky+YIbr7byY1Vav6hq/tRXth5C41puT1Gq9DTNU6QkIzFiLe1I4WGJgTNkSGO9ytN0jvMHlwo+8i3aQTtXr5e9zYAXl1eiCYEhh5G09zs6NjBwfyUFi6yFI2ImoJcM2bXs+1wOPBLPaNup/fcc/H19NBny8doUCUY589/B1decZgXojofEnfxJJeqtsVXAFDZ38WvF5byia4HqDTa2BmfR2c0htusu3vn4Wa+fbydB7v6mYvknCDPtJGP3B7Ya2Z48/oor5tYKPKCYbbzsfI8z1ZceZkeOsMrB7W13c+WLa+kKJ6qtnJLcB6GYdDd/RgAoV6oLfgjVVWvz9h3spOd6cTc3Tik8pQYr4huLE0jj8ooJx7/WFKQ9/VNTpAbUnLn4ZYR6/Nt6hxVNidCg8pXZI4gDCPO22pKabl0FV8eehWB5h8AEAwcA6C7e3NG+/fMK2NNXuZLtD1LrpZ0Yu3tYBhIBIY0iEvwmNGW3mGJsPq0omQE82PF5wJQZioauhHGZnOzo30Ha6pXo3cp+7HMouBEI8rLo/vuQ8T7JhdINt14PJ5MQb5hA5UbNxKzO2j582aVJtHkJbOaUwPLOf+8Rxlyq+/kchnlirp5vPqVn+SVETX/c2j/E7iWLKG9uJQ/ybk9+Zl7gjxtOZtG3rdXuWRVzF8wdoL+NC4tzucLi5XtsNLpQJvgfrOJiJGZX6KnJ1WLMhg8RMPBLwBwrXsfD/QdT7ULpwTkkE1n8dpVVFe9AQBbWNkvJ+pHHgodH7HOcAaxDyYE+djZJFORnXHChAh1rMBuPqID/ZPTpJ7rC/Joz0gf70K7Oodf2PFftQB3WSkXX5T67hIToYlMgUeOfguAE/uURtvT8ySxWOo7+/ziGv55dlrueil5YWBsH+ZYk0rYJYVQpcokuGxKOHuGaeSLZAM3BlNeMf9emc98j/LoMYwhdGy0h9rZUL6BeQteq/rYlumVYxgR4roabdl9Pmz+zALeM43H4yEUyvzOqs5TVZI+sur97H70r1n3e2TQy+UvKI+rV69Ro+njLUMMhNX19T3xLcTAIT78yS8Q1pRyNjhHKwflniBPezsfGcoUMHogQGOPeoirFtRP+JiaELxrXjm7LziLhzbNfIrPk2HevHdkfD5y9DvJ5e07bkwG/Ng0G2ctXsSnfv7/AOjsV8Wng+3LKF3rxWa3UVi4iSuvOEzJgEprMDBOQi5Qpptnn7si6zZbVyEgiEbHrr6T8EKyaToBlLeLzXx1741N7gfYlxZxufSwerkXO2zUmqmMI9jxX6JMKE5nyn0zYUYanq2xsj5lWx4zQZkQvBQIjZlSINas5iYkgrgRJy7BKdRP0ZtmrvmkvINP8UXOHfo9AP54gDuWp0aZuh6mPag01A0VG5i/Ul1Px4nMkU/Chxyg6JqlGcFes4HhGjmQzIn01LpzeLk2n//aowR2WrYVvn4s9cI6t04Vy3DYNCJD6voaXWUEf/l62gtSvvPdp5AbZzaTc4I83Ub+dG+mzTW8ezfthcoeW1JZxWQpdzmocOXmEKyk5BIANKmiWWWaO1e6n7RAw7VkMf6QEpRd5ux+9/6rufbaCzOOWVqnjtna+Idxz59uyuk5eAWlBakXiT1SiEAjHG4aM7goIfw0DIZs6iVtNycX/1E+uYx9HWk/2GVHVc6dxV43/Q8pgRCu9I8p0IbPC2j2lNLwws5beOTRRXR3PzFivxrZiI7g+c7Rffljzc2gaaZpJYYuwaEJYrFeQv0pb6KV7MFHkMICP4+3/4hnDn0u6f4IytuoLRDAhoclhUvw+NQIamBgL9FID719aqQxcCyVvthVOTsCgtLJJsgXeDLNnT/sHCAUjfHphkSVo3kY4dQ+HvNenlNfTBFD2GWM/1z8IT5U9wEA1ja8hENA9yl4P81mck6QJ37sa/I8NITCGTk4Bndsp9MMy3douWfnPhU0030tL28ptmg+Q5HjtLb9aWTqWKGhud2UmPlk9vaYOSmKXJSUZ+aY8S1TLnpDsWMjJk+Hk+5aaC8VrF7/8uRnp7cCW0wdOxLJXtEI0ot3SUK2GMIGtrRAkcgkPA7CphfIdY/cy0UNe1md5+Hjmo9wg4ru9NcUjrpvPD6YIcj7+rYlfbbD3SnvkJ273jZi30+jTFg/3/HiqMePNTdjr6xIauSa5gQZ58U9H2TgyMeT7Rzm/EBpWQHLB/dR6ssMr4/GhggzRK1nBTbNhmZGh/qqn2bL1pvYseNNDA4c4MCJzwDKxXG2JM1KJ5sgX5vv5dP1lRnrvn/fA8nlPM8mDhkp8RUdUoqETRMsLvGyADUX9PcyNSn6lr//kUohLUE+K9DjGEElCNb5vegS9gVSD8Cx7VuJmeG8jlk2fJxuEsFBwi7ID69HSAcvvfQJWloyiwwI85aXVam83LuOKw114/lnjzymPSU4jxz8dsa2np6nicdTwT4dXanh+6o1y5JCBSBvQR3edlXMIDSUmmwdTsIaIZBEHAJfgQtbmrtf1+DEU5M2tau0vK+7+D5Wz1vAw2cvY/WxEDGvei4KK0bPOdLf/wJxPaWBb99xAy+0PkWf4aZZyz7p+u3aFt4kf0kxShN/YXCI1taRKYEBos3NOKtrkIBuRHHavBhGjKGhJoro5VW727g1LbmVy1Whii77M9P8h6IBYiKSNVFWJKryyuzf9l/EnarPPt8StCweLTON3+/H4xn5gvlIXSXnuVP3/38qU+bSv7ZkplnQ00xpy6oqWMTB5Ofr4h2sb3iJ8u4TNA5NLm9QrpBbgvzvH0d/4KMArM5XN/6loJqBl4bBifYWNNPjxHZmyXGEaWOV0sBXWo9EJ9+1moaDd2RtVz5f2RTDQQdSClasH5m+IC9veXK5pTn1QghH2nhh5y28tO+T5jkl//rrn5LbPd5Ms1b+8sW4gkq7T3jIZCMRIyCQODx+fAUuSAv6au4Y2+slnZ5BZU5xOaLUX63qPxa9djFH8tTz4fKMPuEXCB6koSez+k8o2k1cuCgszZ49sLbzK7yKP6n+S52O0kpGy7sfa27BUVOT1MhdDh/SiCajOy/q6eaOcxamSrJJQ5UdLMgU5JHYIDEJr1g0erRmtD9l4vF4ZmdR4ssvv5z3ve99WbfV+tX8hRg25xDyKBPqO2pK+emqOrz+1Iu5siCfN/Ib6o4founStfzoyquwO20sPL6XAwOW++HMs+ltGFGlgc83lHbWFlG24PDhI7R5HFRUKoFhz0HPk1PDvJXSoHjReaAZNB/5N/qNzMRaQtiQUlK2ZJFa4fAijCIcTjfDEcLG2Zv+BEDc6EfX1UszGlWjokTY91NP7Gb+0l8l9/MPKyXmLa6hwLcRgJaW3496Bek/Vbe/HF+hC7eR0vqbuke3Ow8nhhO7jCEAlzmBbRiS9lYzWdUYmun3HjtKT3+mhrimeDHLStayvDhVNUqLe5CmF4TbnRKS5Q6DMAJbluRrMhol3t6eJsijuO15GDIKUvWpfn0eDpct6UFjRPoBCf5hUZkyTExqbKwcPUTfsKdcDV2uilHbzVY+taiaf2/ay8YulZysaKCP17yk4h8cwJ1La7m2rDBjH83mxEOYvoOS9v4BhKbhWrEKz5BQqR72/PE0X8X0k1uCvGot+suUhmn/52fw20TSO2Fwx3ZqewZZcK4Ko3acYYI8P38FRUXns2zZF3B6VeTbnvjjbD5aktFuYGA3jz62mJ7KDpzRGLpDx5c3uqam6jsKpBaj46AyV0QjalgrZQwpJX8f+q+MfXw+5Y63ft0vqap6A3Z7IaUrzgfdntXHOYkpyTUM7IXrlWnF0PhUx9cAONQ3sVwmADHs2M0MeIZTPSPdTQF002tFG6N8XCA8xD1bMn3Q4/F+7I6C5KQy/WsASd8TTXR0PJQMgQdY4VGX0pwlVD/W1gaGkRTkIPE6/UQibQwF1UumbPEJWlruTfrvG1Hzuv2p+ySlxC7iaBTitqdewuXl12acL+pL1XgpLBhpPpvtzHM7+cLNN7FyrUrPW5Dno7BDTXieW5iXdZ/Evf3P877Bg9uU0HYtWwZBnahwIj0lWfcDGBo6wfETPx51+2wltwQ5IMuUkLAF2vCGe/lxUxeGlOh7X+KsYIyqc1UEX7Z0tXMZTXOxYf2v8PtX47AXAnDBgkd59SIVGReOq2H7kWZl2jja8mnsSHS7gdtbnPWYoEqQ2WxehO6gvVsFWkRMQR6NdvHQzo9SFk+5451zzt+w2ZTZorj4Alau+ApCCDyrS3FEisf0SU+JeEmtvx5vgROp2yjX1OirOTjxHNMxnDhMQZ6IKG052IcQunldowvy6xf/ibdvzExP2xvoobnfTlXl67nk4u1ULrwEwx7m2KHv8eKe9zEUPpFsu9qn3kjZoggTroeO2lokyozkFWZBYU352bd0fIt9+z9FOKwElhEz/eHTNPKG9nYcmqTQlWnGWr3qu3jcdVmvy+OZP+o1z2aEEPhMt8ySAj9rq8q5K8/gD+sWZW9vmqgqfJ3kRX/M4OBe3MuXYQ+FkEKg118y6rleeOGtHDr05QkVC59N5Jwg1021Tbvqv2hzFALw95ZWhnbuxLNmDWFTq3OeYYI8HYdj5ESeKP4SAC6RCspxOgIYmsQxTrEBm82HJ28evbEn0fUQkWiqeLaj96+s9aY8S/LT7OrpaG47LlsFCElkKPuEoUzayGFBwSJlI5caxXlOkJL+0OjFm597/moaG+9Ofo6nC3JTELYe6sNbaE4Kj2FaEQJERLkWFpWqLJBuW4CtjU40zYbDUUhhqfKo6Fl874j9N/rUeR/oHEOQp2nkdn3skYaemFROs5FvP6Du5xLvyO9EhLLb/+12f9b1uUBihH1+UT43vec9vPrsDaMG/KWPtso8bWzZ+mqGFg5hN0fvo6W/BpLPtpT6qG1mI7knyM17YKvZyGK7ujEDj3+XR935bL3ocnaaUXUr82afm9XpwuEYqWFfsnzkZKZDxohpArtt7NJ2Pu9CDBnFMIbo6nqM6Cih9nZzJDAa3gLlC9514Mms242k14pBXeFSfAVOpNTwuj3YDZ3+LKHvcUMSMQyCwYPJiV0pJTFsOM320XAPUkpaDvVRuljl5R7NtFJcfHHG5/y00crNl7w2da1ZXpYJljmUAN85GBpR/CTa3EzE7WavLx+JhkOG0eTYSckMfQiceeBKCeKugPI2KvFm9iPWFsTRkj2GwuHIXUGeyBE0kZF2NkUmUhTEoZvBXmOY9xIjxskWU5lpck6QJ/zIhYC/n6/sZgf7JZ9978d5/4KVdMXieG1aTia+mioSninp2GyZ9sR9wUocIkrU5sA+5GFfYIjvHG/PWnWppPRywuEmbLY82jseIBTOLsjXrftp1vUJimrUhGdP41NZtyfOLIBSTwW+QqWROzQHdkMnIEba2F/9wkFWPZVZKUfv72AILx5zclaKCP0dQwwNxhAF96lziOzPh33Y95TI1Q4wP83NT2T56STzuYePUGC3ETEkz/dnukzGmpv53etu5JqdShDXy0PEhkYvFyd0F/GhIDKvOiMY6IgZ0LV65deS66SU9N5/CLvMLrBt47ywZzOb/MpbaIVv5KT8cAoKRlbOChttuPLUvR1LI08Y+E6mTu1MknuC3PxvEwK/3caqPA/bKlM2r45whIIcTHo13Qwvuru0eBNOYsRwktdeye0NTXzpSCuP9owc5vvzVwMq6Odf+zp48cRh9gUyBZ4Qdgr8Yxc4yCtUZpfByD6M6Miha0qQG2iapmzkUkOPx3EhibrcDHRnBhTtGAgxqBtEUUL0yafOp/9H72EAP3YzKCiuB2g51Desv5mP/rnnPMjy5V/KENwAHrcyZyxd8rkMH2yvb6R91iZ8CGFnMLCXS8zasP8cZl6JNbewe0kqqKje3k7Bvuyud4ljSpueYR8Px2KgKROAy5UZNJN/SQ0Vq64GoIDv0N1wJWDjissbZqUP+US5osTPU+cu5zXlheO2dTqLKSm5NGNdS8vvcMxXL7Jsykos1s9L+z6V/Gxp5NNIWDdoMd0NE6K6yuWgwZsKOz7Y10N+jtTcnE6WLPlshmYyXCNfWXcuDqJEdSdifx+t5veacOdMJz8/ZZb5zo634Xf2ENRiDHnPTa6/4PzR/cMTJGy0UW8r4f0jXQmToy3zs9vnQOou4vEgXpuNuN3Bvj3ZIya7KKWNKvIa2+k7soMBCnA51cvr0KGv0HKwA19xSssanh0yL28pNdU3oNkyTXIVFa9izZq7qK3NrEyT51vC6tXfz1jn7J6HlHGikU7OPaTSGjzY3Z9R1CDW3ExpmqNlnvDx/IFFEHxV1uuyez1ITUcUpTxWHjmyC5cWQyKSvuegJgU9Z5VStfEVXHbpiyxffyX+sghOe1UyYCyXWex1TzgR3qqzvkN11Q0Z65zl6t5GwmHaO/7O9h03JrcdPvI/tLam5juMccxds42cEuSfOdjE7WauhYStrNRhp9+V+vE16A5LIwfmz3sbmzamfLY1zY5rcUrw2O35OIgRxUnX3j1Jv/tsubQNfDzWdA0AP3n5h/E7B7l2yY1cc84vk22czuzBL+lopulBd/dhrxs5IReJKUErEpOeQoDhR5d9FLjd6EKj4WAqYi9dQH5CfJfbxP/x7kgpe9/gp1NUsLIkJeS62vdQfXbKf3i45p28VnNInfBqEcJBWemVWQWhx53yAvE0/IiC42pkGA23srpL5UzpD/Ry0ExsZZg+5IZXnfvf9J/hDRWixwzWrv8il1y8bWR/9Ci6MDJcDx85+jxODYTmGVWw2Wxe8orclC4I4svPXg1pLmO357FkyX+iaR50qb6jUKGa6A8eO8GePR+kr29L8n53dT2asf9kiqnMBnJKkJc7UxNUiSxxxaHMgr4RQ+K3BHmSTRvvY/EiFYF5975U8iu7LQ8HMWLCwZHWEwyZtudsuSicdo3Ll2VWWfK7K7FpNjZtvJdFiz6ZEZI/GintUTLEyFD9ngH1Q3N0pxJkaRQinMfwOgU6kqaOTgyzr4EsBZnPD5RwF+9Xx9GP4vWqAJ5orAW3X9nMbTYvfv+aUfqoXjBFRecjhI14fHSPErc75UVy/rsvx1+iRi5RfZCz33wXAAP2PN721BP0D3QRb20FKQl5faz2ObjC+Btan4eqxQWU1uZltdsbsX4MTWSYVvZ0vYAbB057dj/qdIbCzRnBSmcSdrubmpo30ShvASBuV89c8Nix5HcdjSjT1/DMnLplI58+0jMTJgR5TWvTiHaeMy0+fwwKCtaxYMG76RrqYm/3XmK2MvLyVmKz5+EgyjHbIt5529eSJqvRkgpduvETrFzxdarMXOUJ96yCgvXULXj3hPqSLuwHB/eO2N7Tp/zT3WneGXaX0rr1oYMYNhv3rTyXG7fuA6B5sHnEMSpWlhBECbjmE9/n222mu6pjiJh4BoDzxzAD1dW9l+Kii1iy5LNcdOGzWT0gEjgcflYs/wqbNt6LEILyV5wPhoYeDyHc+fx8VR0Ah12V/OHPX+KRF7fSm++nz+XCRQS7tCOjGqsvU4I2m2+7bkQwNIE0J4qllESNPZyXP5Ss+jMahhEhFutO2vnPRJYu+SyvP0+leHaihHNPUxOapiZNO36zha5tz40oyJ1rGnlOzX7UpAlyn2kHP/eFLXDRKzPaxceclT4zKfWU8vDrH0YTGl6Hl3C4FQfPjWg3Wr5mu91HVdX1VFa+jv7+7fT0PEld3Xsm1YeEtqtpLgJZcnoPDHaBrxqfLzXnYXfYkYDdCGA4C2nNK6I5pH50LcE2wEmVbKZVKGH1Z66nhVo2yK28Z+17+MH+v6kDOVL+1i7n6DVc3a5K1q//+ajbh1Nd/YbksrPSj2NPMTHRxVBrOxtLU+amr1X9GwM2H3zth2DAZWIQu+5EOjQWrlftsmnk0nS7k+VLEcChnhN8rrZzRLtsaJqLyy7dc0ql+uYCfl8VbZG1zHepwid7Boc4z+ZF1wPo9iCdTx6EYfP0I7KGznJySiNP9w1P5B/2bN2CLxqh2uXg3yqUAIhmmZW2gDxnHl6Hss+6XJU4ReYPvMhuo2ecNJ9CCCrKr6O37/lkhOdESbjnOZ1lDAZGCvJgQAUKudwpNzm7Q728nYRpzUsJeCkl3TE1KpjPseT642IhMeHEQGN91QX8/JpfAxDydTGoQ5s9rZrPNODxK7t5z8O7KHHY+JpZ/3PAluk1VD7QhF3aKar0YTNHl9ns8E5nOZCy3T/Q8PSk+qNpzlHnA84UhNCYt+iH9IWUzOjt78dmUxq57+XFDFWqkZ3nyG2UlSmPn/GqWc02ckqQV6TZyIUQ6IEAkYMHefTgdp4+dwW3L6xisdfFJ+snX1TiTEMIQV11Zm3OfD1GaAI5vysqXkVl5WswjMnVfkyYVsLhJgYH94zIcR4ZUvZomztl+/V7Ve4Q7zAf6J0nmvjgMfViX5DF3h7HhiacGFE1Cli2oAav3Umpp3JE26kkL38xAIGOQ0QbB3lLdQkuc0KyJpjyF695ogGEpLgqda3ZJi59PvN4AVWcWQsvJ6ybybXqPzw9FzEHuWBRCf0h9ZKNaXE0Qz2Lra2P0VtxPwDzDq+kuFMJ8r7+F2amoydJTgly+7Ac4+Hdu0FKStauwWPTqHU7eercFazzn9kayERxD6sPaXR3EZxA4n2fbyFnrfzGpHN3CKEltU4pdQKBVKpYQxrEwmri2hFKmQ6Kis5Fj7lwk1kx5iOHUxrTAtSQWaSFVb+KP+F0ltB+JIDUbfiLwSajLCrKnkJgqsjLUxOetiuGcM33I4Tg6tICNOCHf/w53/vHvfyxIoa7+xwMLY5jnMA1f76alO3qfgSAhlaQ0klt7S0srP/QtF7LXMJu0yjOX45dxui7ykYwrF7+zUeVO2t5yZvwrqnEeFiNnDo6Hhj1WLORnBLkwwnt3AmAZ212DwSLsXENezF6o2EC01xBJT2HRVfXI7S1/YWBgd20B9uxx83iy4f+lmyjcpLbsQ+bfDos1aO7QB5JFhHYSMp97+radbhc5bQc6kNKOz0BlWY33e96OigoUMbWsCs1Efvx+koe2LiEgkMHOV8z8HdVY4+7wG5kRGsOZ/26X5KXp0xB3V2PA7CrsQ+Hpo+ZvdEiOyvnr8FOjDh2EGrkKewqnqF+8a0U37CMvI0pD59YbOLZNmeanJrsBPjF6nqKzMKsQ7t24Vy8CJs/d3NIzCSuYRp5frCfCLXouo7NZqMtEuM3rd18dEHFhAMxJsPAwC6OHlO1PR0Vb8dpWnXs3lSaUV+BE8Ow49FDqUghIK7ZKBBRviA/jR2dL8uPUkI3G+UW6jiKz/dOQCXKKtyU5ko2zYExbvcCACKRVBrcJT43RjTKgY4OHLU1OL0ah0teYLFNkHFRaaxf9wuKiy+go/MhAALBfbR0N9HUO4RNi037C2kuctb81djbO4mliT1Xfp+aTLf5EJqg6PolOB+tIKq1c+Kh37Hw2n+fdcWqs5FzGvnLSws4u0ANf5w1teRfddUM9yh3cdsyb79rsA9DaBw4rOp4vmnXYb52tI3jWfJqnyzONI+RYCgV3BNr/ykOM3DD5itP9dHnAMOON54ZLwBQIwewk5jwPIGPIJfwOPM5jibsxKI6Hccy85jEYt0jjjOV2M3c4JFhicXiLUqwO2pq8C03eHjp3WhCY7ggX7vmR6xc8XWKi1UhbJuWyi3y/N6f8KWLvoDAQGijVziyyI7PtxAHMeKkRjOSVGwBgNAEFQuuAwldg/+i994GpD77nSdyTiNPp/I/PzfTXchp3GmaxuLGY+QPKWH5zz37WLp4CfvNMnr6FLpzbtxwD52d/+D4iR8THpZ8y2WaS+y+1ISk0AQOXzc+tgCZmQmJBGEMxbT96ADGMA+mSGRirnsnS8IzJ1FFKUHUTF/rrKnhYEAJdS3LKKe09IqMz1qawPbHf0Fi+scyrUweTXPhtZcSio+cQ7PZvPy9sw+nphGM15AnIFj+IkZPBISKeC6dxYn4ck4jt5h63lpdwi/+8FPObldVbrZ19PC23YeT26fSndPrXcA3nz2HI/11RPTMx89dqP7b8qpH7JeP6dGSvjI6eqEJIewMdA5hs2eeI8+3ZJQ9pgYhBELYMYxwhjBPz0PeHFDLGhpiFNNKAm0UzdsyrUwew5A065It4gJ+wruS612uKqLSztv3HOPm3Ud4d9ta7uJ9DAnBX1e38NOWLlY9vYe7m9X9lFLy/peO8+97js3QlYzklAS5EOINQoi9QghDCDEyd6TFrCaZbVAI3GtWszaiNPCQzU5hIDXRE5lCQW4YkvaBCMd7vTi0YROrC5SRfH8gPyOPCkA+ykTismksNAsnu8zERobMFIZ5eSsoK3sZKy+q5p3fvJjqapU8adPGe6mre/+UXctoJIRvINCQXBdrbgG7HXt5Oc2BZgRCaeTjmF8tQT51aGkj0EfFK3B1Xg5AJB7ixDDz4WZxJX/gJm5vreQzB9WL96tHlKLzSM8g97X38tfOPqITcNc9HZyqRr4HuB54Ygr6YnGaSchKDai+804u+P7/Q0hJ3OGgry01WReZwodV0wS/fdd5vOncJWgiU1gnPv1kt545uRqrT2rkAljoVcLN5zCQkhHHWbTo49jtyu/c7rCxdMnnuPii5ykoWD+hnDCnis3MoBgMpgvyZhyVlQi7nZZACxW+CjCLvY2FJcinlnqP+t6Wdh9mXuMiJFBTcSO/a0jFIlQY6ln7h7gOgMt6tgAQiEV5tOE3HAym4icap3D+6FQ4JaOPlHIfZA9ksJj9pJdWA7BpGj6pMxSLog8Oguk8Mh2RstlymEhTr/iP12XmknbzcvL4CQA3VRUnk2X53a3EY04cjswfk2NYSTObzZMUrqcDTfMghJNAuiBvasJRo9IINAeaqfZVA52MJ8h9vkV4vQsJhY5knuM0Xs9c4on1i7h150G2B8s5d+MiLjpcxcGBebQXKM+mW5/+GyUuweaNxWwT57FINvBa1x6WtTTyw+p/I9J0B98RS8CcMO2Mxlk0C8JWTpuNXAjxLiHENiHEts7O6Z1wspgYtW6lnSzyprS+EpeLqBAMulMh5ZFpyF2TXbAqoTa/tChjrdPjxkGc55eW8PnFNRSb7qdOguY+mVrrTNemdDgKkTJKe3vKHz7W3IyjVgnylkALNXk1SCnHtZEDFGapeGPTLEF+MvzmMx+j9YWt9PnUiO2pRetoL1Aayxv1QT500w3cdtuncJveLF5COF2HWH/lRwH4KN+jN83rpXugi9nAuIJcCPEvIcSeLH+vmcyJpJR3SSk3SSk3lZWNn7vaYvq5prSAe9ct4taalEtgiceFXlhKh78IYQrw6bADOszC2QB9jqWAaWiQejIzXQKXx3QNC4SxCZEsxKth4PF4cLvV81RRfh0FBRtmvFp84vy6HkBKiRGJEO/sxFFTQ8yI0R5qpzoxoTuhGpRFI9adzhHGXOLc629g8cqzRqz/r0XVfOeqi6leuBhNs1FdooqmeAkSiXRzQaF6BvuEquH6bz07EdJg6/H9I441E4wryKWUV0kpV2X5+/Pp6KDF9CGE4KKi/Aw3uEK7jUMlVfR585Hm+v741FcUL0krcrw9nM/TrSVINAQj7cJur2lzHlD1LxO9lQji8T7cbiUUfb4lbNr4hxm3H3vTXiSRSCsx04fcWVNDW7ANQxrU5NUwERs5wLz572N316rMlaPUHLUYm+UXXIKtJBWncKOmMmneUl2S0a7Iq+6hk3zKvV1s2Xk3/y0/ndxeuKeDdW6NF331p6HX42O5H1pkkDBbADjj6iF/pjcwWvOTxuWqSC4fGmzHFy4zxZqRzEyXwONT2tDQoPJzX2IW4K1CeRP4TJdCyewI3HC6UoKisemXymMF5XrYYvqQJwT5REwrh7sM7m24btja2eEtkYtcl1b381OHXmTPhavwDStGcyikbObn+dWz7wl9nRoak9vdHe2UHjpMc0QpOY91D3DH4RZmilN1P3ydEKIJOB/4mxDioanplsVMUehIPdDvblYFHJ7oHb3K+6lQXfVGQONwoI/SgHdUjTxhWhkaVH7jrysv5DdLdS7gKQAWzH8PixbexoL575yWfk6W8rJrkssnTtyFo6qSkn//d5z19bQEWiiyGeRF9o9wsRyNXU19NAeqmb/sbyxd+nmAM7pYxKlybWlqot1oOJA10Kc9qpSYZZ7UCzNhNweoufxSSl7aS/NQGENKbtx9hO+d6Eje04PB8JR6e43HKQlyKeX9UspaKaVLSlkhpXzFVHXMYmYotKuHepnPzZsvuwSbHs9ax/NkkFLyq5ZuhkyvkxUrvgz+C4lKg6qgNDVyOSIvt2ZTppJwUGnkQgjkgTchAJstH4+nmrq6980au7HLVUZR0QWpz4sWUX7bx7CXlNAUaOK/qsO0HvlvQE7IRr6rsY8Cj4PF1cuorbmZSy5+YcbnAXIZIQSfrq/kozueInygIWubbyyfx2vKC7lyya34C5UZMP1OLbr8MhZUlBETGpu37Eiu7woN0ROLc/GW/bzvpePTeBWZWKYViwyqzSpM76wtZcGCBRTEouiQFL6nwmM9g3z8QCNfTBuCluWv5IaiKPUDMdPQMFJLTdi8h4LpkZyqP2739OYXPxmklLT3Z8/p0jbYmNbOYCI28p2NfaydV2hGjQocDitJ3KnykbpK3kaEyOHDyFhsxPa1+V5+eFYdHmceK5Z9ZsT2Qnq44iZVHer+f6XCaJ7c9SJPmabIf3T1T1PvR2IJcosM3lRVzGNnL+PmqhKEECx0Kw398eb2cfYcn4SIbgilDVFLz+Pa2k34e3uzCnIZNxCG6kM4NLL8VjaPjpnmcGeA/mDKLU1KnZf2fYpDh75KR7ApbX1sXBt5KBqnoX2QdbWj1w61ODlcS5dBLEbkyMjCJOmkp3W4S97Ch/zb6dx1JbW2NjSgyV+Y3H7fkUZ+36pe4gvcpy+xmSXILTLQhGBFnicZ5PWmBara0o8PN46125h86XALnzrQSMwMLBpI84IpKbmEDRt+jewfzCrIBzc30fsH9UOLhFOCXDP9qCsrrz/pfk0Xi8vzKfOmXlZH216itfVejp+4i2j4RFrL8b1WWvqGqPC7WTuvcFr6eibjPussfJdcDPrETYc+gpzT/2UE8NAP/xtnOMSBmlQO8xfzi3nM1MS7AgHkabKTW4LcYkyurFHeJQ1D0QlPzg2nKxbnTx199MXVD2ZE7hYp0QeUIGeYIHfOz0fElUZeVf8PNm/eAKhqQwDFReeeVJ+mHZlKu7tvzw3J5VsKWjPbjWMjX1yez7O3X8kVy8vHbGcxeVwL65l/1124V64ct+3Zm/5MMKq8pRLKxrLzL6bIYaOvULku1jUeocNfjKFpFAT7CQgNoZ0eEWsJcosxqXQ6sCMZtNlpbDw5rfyionz643oyLe6I2fxQD0ZYxxawYxuWC8W9pIjSt6gKULJ0P3G9n8GdjRhGBCEcuN3zTqpPpxO3LTLG1omlt7DSYMwsfv8q5lWek7FuwZrVVBQWoZtKxfn+VP3Vs8pKMTQbxjRERWfDEuQWYyKEoNhhJ2Z3EgyOLO4wES4qVA/4s31qEmhE6pbeo+gRjbxjhdiH+ZAD2F2Z67r+tBVpxHHaS2etgCssPGf8RoCU01taz2LqWLvmrozPcT1AUZq77upN65PLskfNkUxHnqJsWILcYlwWe93omkZB/aIR24Z0I2lyMaQkmCUKtNzlYKnXzYkhldyqKzrMS6DnKHpMIN3erPqpGBap6bhcpX/VOvwMPtWMPE0/lsmwbu3PuPii5/F6F4/ZTtdP7uVocfoZnjmzP9hFni0lyJd4UwpHtF0VB49ZGrnFbGGdWZYmoVED/L61h4/sO0H9E7v5f40qCdq79h5j0ZMvjrClB+M6moA+U8gHDcnOfftSDUyNXLh9ZCuPqIlMQR6uVZkAPfYF9D9whKE9syNxUTo2mxunsxSnszi5rkNbOqJdPD71UbMW08d55z5Mc+EXMSQ8fujJ5BTHIo+LC4pSphV3TCktU5nLfywsQW4xLsvzlKaxO5DyGvnw/hP8rk1VIP+C6Rf+QKearR/UDR7vGeBf3Sqv8y0vHmV/MJwxjVlQnvL/lt1H0GMa0j3SrAIjc28PDOwCoHTDBZS+fRWe1aXZdpsV2DT1EvT71xKQI99SuiXIcwqfbyE3rns9g7pGd3Avn6ur5La6Ch45exk2Ifjuivl8ZEEF8yqVk0BgaKTL7HRgCXKLcal1KUG63xTkup5pPhFIjPtS4fHb+oO8adcRbt6tNOfmoTDDceel0uTKjiNgCITTmdWverggDwYOAOD1LcK9tGjW2skBvL6FgEpF25elNN3wIs0Wsx+H5sCwFVPk6uWR7X/mE/VVyULmb6gs5tMLq1i+SJkhX9iz97T0yRLkFuOSbyYUOmgmEupqbMrYLhEMdB5Lfr5pd6oIwub9DXQPpHK12EyZG0sbcuptx5FAzOXKaloRwwoNR2PKlGLLgSo5C+s/AmgIzUVvdKT2XVt782nvk8WpU124lKVunXnx/6DfHCGmU1miTGo7du8eofhMB5YgtxiXxIROSyRGUNeJV6ogoc8OdvCpP/0UgGc3V2Td986tuxl0pXKgJOR3chIoGkLv6+LLb30vf3Rnj15UoekjhfbwnCyzEbvdh9e7kL7BAyMEeVXl9VRXv3GGemZxKhR4Un79kfDIUZXD1EgOSxtf2rGXfYHpNbFYgtxiXKpcKY34pUA4OYFTWTbE0kVK+35glfL1LuvJfKj3VtUBsCHUC6TCfZKCvPcYekTj4fMuAVTprGxkyzGeC4IcIC9vKY39+xnuwKDZZkGNMIuTIn2S+nifMp+kT/I7TXPfoys28r2AwUf2n2A6sQS5xbi4bRp5pg1w12AoGdDj2vUrCsuVDfhwmdLIr971NNVdrVzx0jYAdJudjV4n97/iYp44Z1nygYsnVPPeo+gxjZqOYRGPw8haLELkxuPrz19Fn+HBNay7dkuQ5yxFhRcmlxvaHuGRRxexbfvrk+vyh+U3bwyNnCeaSnLjl2Ax4yQqqOweDBE2C0640Cm89DYA3JvUg3375/6TJ1bX88Vldfjt6vFaU+THZbez1OdhsVkfNJrQXnqOYkQ0ak2/22OXrMl6/uwaeW5UyVmw4N3k1bwTl1DXbLcrE5Ivb9lMdsviFKitvZnokR8BUBlXrrQDAzuT21fmpcyJlYO99OgyY15oqrEEucWESMzK7xoIEdl1j1p39jsoKFL28sNDETSgwG4jb/lyFl51FRv9yjNlVdpDvTpfLf8zkeKz9yi69BB1ODg7z508z3CGT3gCiBx6fJsHmzkRVS+jtWvuYsOG31FZMamytxazCE3TuOyGizBi2V1m/XYbj529jMMXr+Yqr3p2G7uzpzaekv5M25Et5hRuM/mPs30PYVOQuxZdSrHDjkMIOqNxFnld2NJcAd83r5zzC31cVpyfXLcuX5kTvnW8g0BcV1Gdooiow4nHMbqGnbCH6zHXiHW5QEughX57LZdesovCwk0UFZ49q90mLcbHk+/E4czPWBdLK4u4Is+Dz25j1cI6AB55YRefPNDIweDUm1ksQW4xIVzmLPxXDn6DoEcF4LhtGl6bxqYCJZwvKsp8qC8uzuf+9UuocafMIivStPPj4ajSyMkj5nbjGitTnFR2+dhgyjtG5IiNHKA52ExNfg12e974jS1yBo+vLOPzwPPH6O/fwdPPXEYk0gFAeb66549E4Rct3dNSzDx3fgkWM0pCI/9S/bt4x+JPAuAzzSCvLlfFHd5QOX6RB2+a6aQpNAR9JwiKfBpq5idfFtlIFFaWRmqCMNc0clVw2WIuUTQsOVpPw/Ns2/4GwuFGAgFlO/ebE59NZdXYBWzwT/0ktyXILSaEq0tFUz5dtCG5LlEB5W01pTRcvJoNfl/WfdNxppkTTvS2Yxg6H73qVgCOmUm1slFYuAmA2mVVyXW5MtkZjofpGuqi2lc9012xmGJcrsw88c1L/i+5vP/Af7L7xfey2que+UOhCAV2O9o0mNQsQW4xPvEI7u0/zVj1sbqKZNADpLSO8ZjndlJhVi0/1N3O/827ia3zVIbA2+qyBxUBaJp6aWjpwjtHTCstQZWLpjrPEuRzDaczM8+P7hpILofDTXR2/pOj+z5IsZnu1jZN0yK58UuwmFnsLlznvD35UQAfmD+60B2LAoedreepiiztg708XHJ+cts1ZYWj7pewh6dr4YLcMK20BJQgr82vHaelRa7h96sc5AWF54/aprvnCb60RN175xjmw1PBEuQWE8JdtTq5LIG9pxBy7LRpOISgLzKUTJLlMMabAFKPakIzh9yZ7JyfP5+PbPgI9f76me6KxRTj8y3k/PP+xaqVXxuzXbHpkTXPPT35gXLjl2Ax4wz3KNk1ODKT32SodTtosRci7erBdsuxi9QmXPU0W+65H873z+cdq99BobtwprtiMQ14vfW43dV4PHWjttlU4OPfa0v58tLpGZVZgtxiQrjThoRlTju7T1GQr8v30u0oJG4mw9LGqV1ptykXLoe9MLkuVwS5xZlBVeXrRt3mtWncsaSW5T7PqG1OBUuQW0yIRMSlTcCaPC+7B08tm9tKn5ug3csL3joAChk7fLmu7v3U13+Y2tq3JNdZgtxiNhGPpyY6bdHMTJ6GMb21WS1BbjEhEpM0RXY7a/I9NATDhPSxzSFjcY0vc983VhaP0lJhs3lYWP8h7HZ/+tqTPr+FxVRTUnIpoNIT2xyZofu6Pr2VoCxBbjEhbKbpY73fyzq/FwNoOIVQ48VDjdSEVaIslyb42FkLJ9YPW2pomiuTnRZnBsXFF3L5ZQdYufLraC41l5PwrErX1qeDWRNREYvFaGpqIhye3nSPcxG3201tbS0Ox8jEUlNFvdfF91cu4GUlfuxCsP+iVRSOkRtlXHqP4jJUAFGxwz7hvCPpwtsyrVjMNjRN/SZKyi6mufnXVJy4mbb5PycWH2B6rOOKWSPIm5qayM/Pp66uzkomNAmklHR3d9PU1ER9/fS6t11fkQrB94ySpXDC9BzFIZcDEJ6kiebsTX+is+thSyO3mLUsXvQJSgauprf9JZgP8Vj/tJ5v1vwSwuEwJSUllhCfJEIISkpKcm8k03sUp3mrB3UdY3j5nDHw+1ezaOHHpqljFhanjt2eT+m6c7FpKpFcPD44zh6nxqwR5IAlxE+SnPzeeo7isKkBYVzCYbOws4XFXEE4bOQtmwdANNg7rec6JUEuhPi6EGK/EGK3EOJ+IUThFPXLYq7TexSnKchfWVbAEl/2BP0WFrlM/oZFAIROtEzreU5VI38YWCWlXAM0ALefepdmjqamJl7zmtewZMkSFi5cyAc+8AEiEaUpPvXUU5xzzjksX76cZcuW8f3vfz9j37vuuovly5ezfPlyNm3axOOPPz4DV5AjRAYh2InToYKB8m3WpKXF3MRTUwZSY6i5LaM481RzSoJcSvlPKWXC0/05IGezAkkpuf7663nta1/LwYMHOXjwIENDQ3zyk5+kra2Nm266iR/84Afs37+fp59+mp/85Cfcf//9ADzwwAP88Ic/5KmnnmL//v3cdddd3HzzzTQ3N8/wVc1Seo8B4HCoefxTnji1sJilCKFhF3nEo/1ET0yfnXwqvVbeDvx+tI1CiHcB7wKYP3/+mAdqaLiDQTMp+1SRn7eCpUs/N+r2Rx99FLfbzdve9jYAbDYb3/zmN1mwYAGapnHrrbeyYYPKxV1aWsrXvvY1Pve5z/G6172Or371q3z961+ntFSltNywYQNve9vb+N73vseXvvSlKb2OOUHPUQBsDg9EwTNWZSALixwnzgB98x8hsKUV1wL/+DucBOP+goQQ/xJC7Mny95q0Np8B4sCvRzuOlPIuKeUmKeWmsrKy0ZrNGHv37mXjxo0Z6/x+P3V1dezatWvEtk2bNvHSSy+Num/6doth9CpB/s+gmqQN6lNf+srCYrYhHaMXTjlVxtXIpZRXjbVdCHErcB1wpZwiI9BYmrPFHKDnKHiKuLrUzz+6BgicQqi/hUWukH9N+fiNTpJT9Vq5Gvgk8Gop5amlw5thVq5cyfbt2zPWDQwM0NbWxvr160ds2759O5s2bRp13/TtFsPoPQpF9bzWrPUZNixBbjF3OWvlNwGIx6cv38qpGif/D8gHHhZC7BRC/GAK+jQjXHnllYRCIX7xi18AoOs6t912Gx/4wAf4+Mc/zt13383OnTsB6O7u5jOf+Qyf+5waOXzyk5/kU5/6FN3d3QDs3LmT+++/n3e/+90zci2znp6jUFyfLOgcNaZvNt/CYqax26c/KOiUJjullIunqiMzjRCC+++/n/e///3ccccddHZ2csMNN/CZz3wGgF/96le8613vor+/n2PHjnH33Xdz6aUq29mrX/1qWlpauPDCC4nH47S1tbFr1y5m41zAjCMl1GyA+edzfqGPBW4nH6urnOleWVhMGza7yqUfn84MiFLK0/63ceNGOZyXXnppxLqZ5Omnn5bz58+X27dvH7Hte9/7nly1apXs6ekZsS0Wi8mbb75Z3nTTTdIwjNPRVSnl7Pv+LCwsFP39u+S/HlkoOzoePuVjAdtkFpk6a5JmzTYuuOACjh8/nnXb+973Pt73vvdl3Wa32/nlL385nV2zsLDIITQz9bJhTF8+JMuB18LCwmIasWkq/YRuCXILCwuL3EQzBbmhW4LcwsLCIiex2SyN3MLCwiKnsTRyCwsLixxH0xzYbHmEI9OXytYS5GnYbDbWrVvH2rVr2bBhA8888wwvvvgi69atY926dRQXF1NfX8+6deu46qqriMVifPrTn2bJkiVs2LCB888/nwcffHDEce+++24+8IEPZKy77LLL2LZtGwA//elPWb16NWvWrGHVqlX8+c9/BuDWW2+lvr6etWvXsnTpUm655Raampqm/4uwsLCYUoqLzqen+8lpS2VruR+m4fF4ktGbDz30ELfffjubN29Orrv11lu57rrreP3rXw/Apz/9aVpbW9mzZw8ul4v29nY2b948qXM2NTVx5513smPHDgoKCggEAnR2dia3f/3rX+f1r389Ukq+9a1vccUVV7Bnzx6cTueUXLOFhcX0U1C4ic6uh3nxxfeyfPmdOJ0lU3r8WSnIP3ewiT2BoSk95qo8D3csmXi69IGBAYqKikbdHgqF+NGPfsTRo0dxuVwAVFRU8MY3vnFS/ero6CA/P5+8PBX9lZeXl1xORwjBRz/6Ue6//34efPBBXvOa14xoY2FhMTtxu6oA6Ox6mOqBGygtvXxKjz8rBflMMTQ0xLp16wiHw7S2tvLoo4+O2vbQoUPMnz8fv//U8guvXbuWiooK6uvrufLKK7n++ut51ateNWr7DRs2sH//fkuQW1jkEC53Kg3FVAtxmKWCfDKa81SSblp59tlnueWWW9izZ88pFzcebX8hBDabjX/84x9s3bqVRx55hI9+9KNs376dz3/+81n3mS4bm4WFxfSR0MinC2uycxTOP/98urq6MuzV6SxevJgTJ04wMDAwYtv999+fnCDdtm0bJSUl9PZmVtHu6elJVhQSQnDOOedw++2387vf/Y777rtv1H698MILrFix4hSuzMLC4nTjdE5vAj1LkI/C/v370XWdkpLskxJer5d3vOMdfPjDHyYaVZU/Ojs7+cMf/sDrXvc6du7cyc6dO9m0aRNnn302Tz/9NG1tbQBs27aNSCTCvHnzaGlpYceOHcnj7ty5kwULFow4n5SS73znO7S2tnL11VdPwxVbWFhMF5rmmNbjz0rTykyRsJGDEpw///nPsY1R4f2LX/win/3sZ1m5ciVutxufz8cXvvCFEe0qKir49re/zbXXXothGOTl5fHb3/4WTdOIxWJ8/OMfp6WlBbfbTVlZGT/4QSqt+yc+8QnuuOMOQqEQ5513Ho899pjlsWJhYZGBmAmb66ZNm2TChzrBvn37LJPBKWB9fxYWs5vOrkcwjCgV5dec9DGEENullCNKj1kauYWFhcVpoKz0ymk7tmUjt7CwsMhxZpUgt1zrTg7re7OwOLOZNYLc7XbT3d1tCaVJIqWku7sbt9s9012xsLCYIWaNjby2tpampqZR/bYtRsftdlNbOzNBVBYWFjPPrBHkDoeD+vr6me6GhYWFRc4xa0wrFhYWFhYnhyXILSwsLHIcS5BbWFhY5DgzEtkphOgEjp/k7qVA1xR2ZyaZK9cyV64D5s61zJXrgLlzLVNxHQuklCMycM2IID8VhBDbsoWo5iJz5VrmynXA3LmWuXIdMHeuZTqvwzKtWFhYWOQ4liC3sLCwyHFyUZDfNdMdmELmyrXMleuAuXMtc+U6YO5cy7RdR87ZyC0sLCwsMslFjdzCwsLCIg1LkFtYWFjkODklyIUQVwshDgghDgkhPj3T/RkLIcQ8IcRjQoiXhBB7hRAfNtcXCyEeFkIcNP8XmeuFEOI75rXtFkJsmNkryEQIYRNCvCCEeMD8XC+EeN7s7++FEE5zvcv8fMjcXjejHR+GEKJQCHGvEGK/EGKfEOL8XLwnQoiPms/VHiHEb4UQ7ly5J0KInwohOoQQe9LWTfoeCCHearY/KIR46yy6lq+bz9duIcT9QojCtG23m9dyQAjxirT1pybbpJQ58QfYgMPAQsAJ7AJWznS/xuhvFbDBXM4HGoCVwNeAT5vrPw181Vy+FngQEMB5wPMzfQ3DrudjwG+AB8zP9wBvMpd/ALzXXH4f8ANz+U3A72e678Ou4+fAO81lJ1CYa/cEqAGOAp60e3FrrtwT4BJgA7Anbd2k7gFQDBwx/xeZy0Wz5FpeDtjN5a+mXctKU265gHpTntmmQrbN+EM5iS/sfOChtM+3A7fPdL8m0f8/Ay8DDgBV5roq4IC5/EPgxrT2yXYz/QfUAo8AVwAPmD+qrrSHNXlvgIeA881lu9lOzPQ1mP0pMAWgGLY+p+6JKcgbTSFmN+/JK3LpngB1w4TfpO4BcCPww7T1Ge1m8lqGbXsd8GtzOUNmJe7LVMi2XDKtJB7eBE3mulmPOZRdDzwPVEgpW81NbUCFuTybr+9bwCcBw/xcAvRJKePm5/S+Jq/D3N5vtp8N1AOdwM9MM9GPhRA+cuyeSCmbgf8BTgCtqO94O7l5TxJM9h7MynuThbejRhQwjdeSS4I8JxFC5AH3AR+RUg6kb5Pq9Tur/T+FENcBHVLK7TPdlynAjhoG/z8p5XogiBrGJ8mRe1IEvAb1YqoGfMDVM9qpKSQX7sFEEEJ8BogDv57uc+WSIG8G5qV9rjXXzVqEEA6UEP+1lPKP5up2IUSVub0K6DDXz9bruxB4tRDiGPA7lHnl20ChECJRmCS9r8nrMLcXAN2ns8Nj0AQ0SSmfNz/fixLsuXZPrgKOSik7pZQx4I+o+5SL9yTBZO/BbL03AAghbgWuA95svphgGq8llwT5VmCJOTPvRE3a/GWG+zQqQggB/ATYJ6X837RNfwESM+xvRdnOE+tvMWfpzwP604aaM4aU8nYpZa2Usg71nT8qpXwz8BjwerPZ8OtIXN/rzfazQruSUrYBjUKIZeaqK4GXyLF7gjKpnCeE8JrPWeI6cu6epDHZe/AQ8HLx/9u3Y5WIgSAAw/9W2uobeGBvYWlhJXqFta34GFf5EIK9hYWNWHsvIBaiForxNa65IhY7YpAD4RTWgf+DFNmk2GGTgZ1MSlmLHcpejDVXStmnliIP+76fDS7dAEfRRbQBbAJ3/EVua/nBY4mPCmNq98c7MGk9nx/mukPdHj4CD3GMqbXJKfAG3ALrcX8BziK2J2C7dQwLYtrlq2tlFA9hB1wBKzG+GuddXB+1nve3GLaA+1iXa2rHQ7o1AU6BF+AZuKB2QqRYE+CSWtufU3dJJ8usAbX+3MVx/I9i6ag178/3/nxw/yRieQUOBuO/ym3+oi9JyWUqrUiSFjCRS1JyJnJJSs5ELknJmcglKTkTuSQlZyKXpOQ+AA+mU0+sy4k3AAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "def ticker_offset(str1, str2, silent=False):\n",
    "    a = yf.download(tickers=str1, period=global_period, interval=global_interval)\n",
    "    b = yf.download(tickers=str2, period=global_period, interval=global_interval)\n",
    "    df = a.merge(b,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime\", suffixes=[\"_a\",\"_b\"])\n",
    "    y1 = df['Close_a']\n",
    "    y2 = df['Close_b']\n",
    "    y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "    y2 = (y2 - np.mean(y2)) / np.std(y2)\n",
    "    if not silent:\n",
    "        plt.plot(range(len(y1)),y1, label=str1)\n",
    "        plt.plot(range(len(y2)),y2, label=str2)\n",
    "        plt.legend()\n",
    "        plt.title(\"Normalized price over last day [minute data]\")\n",
    "        plt.ylabel(\"Normalized price\")\n",
    "        plt.xlabel(\"minute\")\n",
    "        plt.show()\n",
    "    return max_offset(y1,y2)\n",
    "\n",
    "def ticker_offset_slope_loss(str1, str2, silent=False):\n",
    "    a = yf.download(tickers=str1, period=global_period, interval=global_interval)\n",
    "    b = yf.download(tickers=str2, period=global_period, interval=global_interval)\n",
    "    df = a.merge(b,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime\", suffixes=[\"_a\",\"_b\"])\n",
    "    y1 = df['Close_a']\n",
    "    y2 = df['Close_b']\n",
    "    y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "    y2 = (y2 - np.mean(y2)) / np.std(y2)\n",
    "    if not silent:\n",
    "        plt.plot(range(len(y1)),y1, label=str1)\n",
    "        plt.plot(range(len(y2)),y2, label=str2)\n",
    "        plt.legend()\n",
    "        plt.title(\"Normalized price over last day [minute data]\")\n",
    "        plt.ylabel(\"Normalized price\")\n",
    "        plt.xlabel(\"minute\")\n",
    "        plt.show()\n",
    "    offset = max_offset(y1,y2)[0]\n",
    "    b[\"Datetime_shift\"] = b.index + pd.Timedelta(minutes=offset)\n",
    "    shift_df = a.merge(b,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime_shift\", suffixes=[\"_a\",\"_b\"])\n",
    "    y1 = shift_df['Close_a']\n",
    "    y2 = shift_df['Close_b']\n",
    "    y1_original = y1[:]\n",
    "    y2_original = y2[:]\n",
    "    \n",
    "    y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "    y2 = (y2 - np.mean(y2)) / np.std(y2)\n",
    "    print(y1, y2)\n",
    "    f, ax = plt.subplots()\n",
    "    def slope_loss(k, s):\n",
    "        diffs = []\n",
    "        for i in range(0, len(y1)-k, s):\n",
    "            xs = np.array(range(i, i+k)).reshape(-1, 1)\n",
    "            line1 = LinearRegression().fit(xs, y1[i:i+k])\n",
    "            starty = line1.predict([xs[0]])\n",
    "            endy = line1.predict([xs[-1]])\n",
    "            ax.plot([xs[0], xs[-1]], [starty, endy])\n",
    "            reg1 = line1.coef_[0]\n",
    "            #print(\"Slope 1 \", reg1)\n",
    "            line2 = LinearRegression().fit(xs, y2[i:i+k])\n",
    "            reg2 = line2.coef_[0]\n",
    "            starty = line2.predict([xs[0]])\n",
    "            endy = line2.predict([xs[-1]])\n",
    "            ax.plot([xs[0], xs[-1]], [starty, endy])\n",
    "            #print(\"Slope 2\", reg2)\n",
    "#             print(type(xs[0]))\n",
    "            diffs.append((abs(reg1-reg2), list(xs[0])[0]))\n",
    "#             if i*s < len(y1)-k: \n",
    "#                 print(shift_df.iloc[i * s, :])\n",
    "#         print(np.where(shift_df[shift_df[]))\n",
    "\n",
    "        #print(diffs)\n",
    "        return np.sum(diffs), diffs\n",
    "        \n",
    "    total_slope_loss, diffs = slope_loss(30, 15)\n",
    "    if not silent:\n",
    "        ax.plot(range(len(y1)),y1, label=str1)\n",
    "        ax.plot(range(len(y2)),y2, label=str2)\n",
    "        ax.legend()\n",
    "        ax.set_title(\"Shifted price over last day [minute data]\")\n",
    "        #ax.ylabel(\"Normalized price\")\n",
    "        #ax.xlabel(\"minute\")\n",
    "        #ax.show()\n",
    "    return offset, total_slope_loss, diffs, y1_original, y2_original\n",
    "print(ticker_offset(\"QQQ\",\"BTC-USD\"))\n",
    "print(ticker_offset_slope_loss(\"QQQ\",\"BTC-USD\")[:3])\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Methods\n",
    "\n",
    "We realized an issue of cross-correlation is that it is influenced a lot by magnitudes. We care a lot more about the slopes.\n",
    "\n",
    "Our idea: SLOPE LOSS. Use the difference in slopes of the regression line in the loss function. Hyperparameters: stride of slope intervals, length of slope interval.\n",
    "\n",
    "Our method was centered around performing cross-correlation to see the shift in time that two assets have. So, once we identify a shift in the earlier portion of a day’s data and see the quality of that shift, we can trade off of that for the rest of the day. Some things we could do in the future are vary the level at which we want to trade and such, but for the purposes of this project, we focused on implementing the strategy. In order to evaluate the success of a shift, we used something we came up with to find the loss based on the slopes. That requires some hyperparameters for interval length and stride length, which refer to how long we take a local slope over and stride length refers to how spaced apart start points are. We took numbers that seemed reasonable for this as 25min interval length and 15min stride length [stride length < interval length allows for overlap]. There is a sample graph for this in the slides.\n",
    "\n",
    "Our trading strategy is as such: wait for the first half of the day until we have enough bars and perform cross correlation to find a the time shift. If such a shift exists, use the shift to trade. If the difference in slopes is greater than a predefined threshold (hyperparameter), and the asset that is leading has the greater slope, then buy. Do the same but in the downwards direction for selling.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "## Early ticker offset does not plot and only uses the \n",
    "# first half of the data to find the shift for correlation\n",
    "from datetime import timedelta\n",
    "global_interval = \"1m\"\n",
    "\n",
    "def early_ticker_offset(str1, str2, start_date):\n",
    "    start_ts = pd.Timestamp(start_date)\n",
    "    end_ts = start_ts + timedelta(days=1)\n",
    "    end_date = str(end_ts).split()[0]\n",
    "    a = yf.download(tickers=str1, interval=global_interval, start=start_date, end=end_date, progress=False)\n",
    "    b = yf.download(tickers=str2, interval=global_interval, start=start_date, end=end_date, progress=False)\n",
    "    df = a.merge(b,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime\", suffixes=[\"_a\",\"_b\"])\n",
    "    \n",
    "    ## Use only the first half of the day\n",
    "    df = df[:len(df)//2]\n",
    "    y1 = df['Close_a']\n",
    "    y2 = df['Close_b']\n",
    "    y1_original = y1[:]\n",
    "    y2_original = y2[:]\n",
    "    \n",
    "    y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "    y2 = (y2 - np.mean(y2)) / np.std(y2)\n",
    "\n",
    "    offset = max_offset(y1,y2)[0]\n",
    "    b[\"Datetime_shift\"] = b.index + pd.Timedelta(minutes=offset)\n",
    "    shift_df = a.merge(b,how=\"inner\", left_on=\"Datetime\", right_on=\"Datetime_shift\", suffixes=[\"_a\",\"_b\"])\n",
    "    y1 = shift_df['Close_a'][:len(df)//2]\n",
    "    y2 = shift_df['Close_b'][:len(df)//2]\n",
    "    y1 = (y1 - np.mean(y1)) / np.std(y1)\n",
    "    y2 = (y2 - np.mean(y2)) / np.std(y2)\n",
    "    \n",
    "    def slope_loss(k, s):\n",
    "        diffs = []\n",
    "        for i in range(0,len(y1)-k, s):\n",
    "            xs = np.array(range(i,i+k)).reshape(-1, 1)\n",
    "            line1 = LinearRegression().fit(xs, y1[i:i+k])\n",
    "            starty = line1.predict([xs[0]])\n",
    "            endy = line1.predict([xs[-1]])\n",
    "\n",
    "            reg1 = line1.coef_[0]\n",
    "            #print(\"Slope 1 \", reg1)\n",
    "            line2 = LinearRegression().fit(xs, y2[i:i+k])\n",
    "            reg2 = line2.coef_[0]\n",
    "            starty = line2.predict([xs[0]])\n",
    "            endy = line2.predict([xs[-1]])\n",
    "\n",
    "            #print(\"Slope 2\", reg2)\n",
    "            diffs.append((abs(reg1-reg2), list(xs[0])[0]))\n",
    "        #print(diffs)\n",
    "        return np.sum(diffs), diffs\n",
    "        \n",
    "    total_slope_loss, diffs = slope_loss(30, 15)\n",
    "    return offset, total_slope_loss, diffs, y1_original, y2_original\n",
    "\n",
    "def backtest_one_day(date_str):\n",
    "    token_diffs = {}\n",
    "    y1y2_original = {}\n",
    "    tokens = [\"QQQ\", \"BTC-USD\", \"ETH-USD\", \"BNB-USD\", \"XRP-USD\", \"LUNA1-USD\", \"ADA-USD\", \"ALGO-USD\", \"XRP-USD\", \"LUNA1-USD\", \"ADA-USD\", \"HEX-USD\", \"AVAX-USD\", \"DOGE-USD\"] # \n",
    "    offset_matrix = {}\n",
    "\n",
    "    for idx1, token1 in enumerate(tokens):\n",
    "        for idx2, token2 in enumerate(tokens):\n",
    "            if token1 == token2: \n",
    "                continue\n",
    "            offset, sl, diffs, y1, y2 = early_ticker_offset(token1, token2, date_str)\n",
    "            pair_symbol = token1 + token2\n",
    "    #         print(pair_symbol)\n",
    "            offset_matrix[pair_symbol] = offset\n",
    "            token_diffs[pair_symbol] = diffs\n",
    "            y1y2_original[pair_symbol] = (y1, y2)\n",
    "    #         print(offset_matrix, token_diffs, y1y2_original)\n",
    "            ##filter for time shifts that make sense and that we can use\n",
    "            # 15min shift is the limit we picked, and <0 so we can use it to trade\n",
    "            # if -15<offset<0:\n",
    "            #     print(token1, token2, offset, \"slope loss\", sl, diffs)\n",
    "    \n",
    "    trading_trigger_threshold = 0.01\n",
    "    pair_symbols = offset_matrix.keys()\n",
    "    pnls = []\n",
    "    pnls_dict = {}\n",
    "    all_trades = []\n",
    "    trades_dict = {}\n",
    "    for pair_symbol in pair_symbols:\n",
    "        offset = offset_matrix[pair_symbol] \n",
    "        diffs = token_diffs[pair_symbol]  \n",
    "        y1_original, y2_original = y1y2_original[pair_symbol]    \n",
    "        # print(f\"Starting {pair_symbol}\")\n",
    "        pnl = 0\n",
    "        hold = False\n",
    "        enter_price = 0\n",
    "        for i in range(len(diffs)-1):\n",
    "            if diffs[i+1][0] - diffs[i][0] > trading_trigger_threshold and hold is False: # and diffs[i+1][1]+offset < max_timestamp: # buy\n",
    "                # pnl -= y2_original[diffs[i+1][1]]\n",
    "                hold = True\n",
    "                enter_price = y2_original[diffs[i+1][1]]\n",
    "                # print(f\"Buy @ {y2_original[diffs[i+1][1]]:.4f} | P/L: {pnl}\")\n",
    "            elif diffs[i][0] - diffs[i+1][0] > trading_trigger_threshold and hold is True: # and diffs[i+1][1]+offset < max_timestamp: # sell\n",
    "                exit_price = y2_original[diffs[i+1][1]]\n",
    "                result = 100*(exit_price - enter_price) / enter_price\n",
    "                pnl += result\n",
    "                all_trades.append(result)\n",
    "                pair_trades = trades_dict.get(pair_symbol, [])\n",
    "                pair_trades.append(result)\n",
    "                trades_dict[pair_symbol] = pair_trades\n",
    "                hold = False\n",
    "                # print(f\"Sell @ {y2_original[diffs[i+1][1]]:.4f} | P/L: {pnl}\")\n",
    "        if hold:\n",
    "            exit_price = y2_original[diffs[-1][1]]\n",
    "            result = 100*(exit_price - enter_price) / enter_price\n",
    "            pnl += result\n",
    "            all_trades.append(result)\n",
    "            pair_trades = trades_dict.get(pair_symbol, [])\n",
    "            pair_trades.append(result)\n",
    "            trades_dict[pair_symbol] = pair_trades\n",
    "            hold = False\n",
    "            # print(f\"Sell @ {y2_original[diffs[-1][1]]:.4f} | P/L: {pnl}\")\n",
    "        pnls.append(pnl)\n",
    "        pnls_dict[pair_symbol] = pnls_dict.get(pair_symbol, 0) + pnl\n",
    "        # print(f\"Ending {pair_symbol} | P/L %: {pnl:.2f}\")\n",
    "    pair_symbols = offset_matrix.keys()\n",
    "    return ([*pair_symbols], pnls, pnls_dict, all_trades, trades_dict)\n",
    "    # plt.figure(figsize=(25, 15))\n",
    "    # plt.bar(pair_symbols, pnls)\n",
    "    # plt.title(\"Our Strategy's P/L Percent for Each Pairing Traded\")\n",
    "    # plt.xlabel(\"Pairing Traded\")\n",
    "    # plt.ylabel(\"P/L %\")\n",
    "    # plt.xticks(rotation='vertical')\n",
    "    # plt.show()\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_results(pair_symbols, pnls):\n",
    "    plt.figure(figsize=(25, 15))\n",
    "    plt.bar(pair_symbols, pnls)\n",
    "    plt.title(\"Our Strategy's P/L Percent for Each Pairing Traded\")\n",
    "    plt.xlabel(\"Pairing Traded\")\n",
    "    plt.ylabel(\"P/L %\")\n",
    "    plt.xticks(rotation='vertical')\n",
    "    plt.show()\n",
    "    \n",
    "# symbols, pnls, pnls_dict, all_trades, trades_dict = backtest_one_day(\"2022-04-29\")\n",
    "# plot_results(symbols, pnls)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "def stats(trades, name):\n",
    "    def sub_stats(trades, name):\n",
    "        if not len(trades):\n",
    "            return\n",
    "        avg = np.mean(trades)\n",
    "        sd = np.std(trades)\n",
    "        print(f\"{name} | Avg Trade {avg:.2f}% | SD {sd:.2f}%\")\n",
    "    \n",
    "    wins = trades[trades > 0]\n",
    "    losses = trades[trades < 0]\n",
    "    scratch = trades[trades == 0]\n",
    "    if not len(trades) or len(scratch) == len(trades):\n",
    "        return\n",
    "    print(f\"{name} | Total P/L % :: {np.sum(trades):.2f}%\")\n",
    "    print(f\"Winrate {100*len(wins) / (len(trades) - len(scratch)):.2f}% | {len(trades)} Trades, {len(wins)} Winners, {len(losses)} Losers, {len(scratch)} Breakeven\")\n",
    "    sub_stats(trades, \"All Trades\")\n",
    "    sub_stats(wins, \"Winners\")\n",
    "    sub_stats(losses, \"Losers\")\n",
    "    print(f\"Profit Factor: {abs(np.sum(wins) / np.sum(losses))} | Gross Profit: {np.sum(wins):.2f}% | Gross Loss: {np.sum(losses):.2f}%\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "history = {}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Ignoring 2022-04-16\n",
      "Ignoring 2022-04-17\n",
      "Done with 2022-04-18\n",
      "Done with 2022-04-19\n",
      "Done with 2022-04-20\n",
      "Done with 2022-04-21\n",
      "Done with 2022-04-22\n",
      "Ignoring 2022-04-23\n",
      "Ignoring 2022-04-24\n",
      "Done with 2022-04-25\n",
      "Done with 2022-04-26\n",
      "Done with 2022-04-27\n",
      "Done with 2022-04-28\n",
      "Done with 2022-04-29\n",
      "Ignoring 2022-04-30\n",
      "Ignoring 2022-05-01\n",
      "\n",
      "1 Failed download:\n",
      "- QQQ: Data doesn't exist for startDate = 1651464000, endDate = 1651550400\n",
      "\n",
      "1 Failed download:\n",
      "- BTC-USD: Data doesn't exist for startDate = 1651464000, endDate = 1651550400\n"
     ]
    },
    {
     "ename": "KeyError",
     "evalue": "'Datetime'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyError\u001b[0m                                  Traceback (most recent call last)",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_5828\\1865177975.py\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     10\u001b[0m         \u001b[0mcurrent_date\u001b[0m \u001b[1;33m+=\u001b[0m \u001b[0mdelt\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     11\u001b[0m         \u001b[1;32mcontinue\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 12\u001b[1;33m     \u001b[0mhistory\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mcur_date_str\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mbacktest_one_day\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mcur_date_str\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     13\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Done with\"\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mcur_date_str\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m     \u001b[0mcurrent_date\u001b[0m \u001b[1;33m+=\u001b[0m \u001b[0mdelt\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_5828\\3365857005.py\u001b[0m in \u001b[0;36mbacktest_one_day\u001b[1;34m(date_str)\u001b[0m\n\u001b[0;32m     63\u001b[0m             \u001b[1;32mif\u001b[0m \u001b[0mtoken1\u001b[0m \u001b[1;33m==\u001b[0m \u001b[0mtoken2\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     64\u001b[0m                 \u001b[1;32mcontinue\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 65\u001b[1;33m             \u001b[0moffset\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msl\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdiffs\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0my2\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mearly_ticker_offset\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtoken1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mtoken2\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdate_str\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     66\u001b[0m             \u001b[0mpair_symbol\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mtoken1\u001b[0m \u001b[1;33m+\u001b[0m \u001b[0mtoken2\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     67\u001b[0m     \u001b[1;31m#         print(pair_symbol)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\ipykernel_5828\\3365857005.py\u001b[0m in \u001b[0;36mearly_ticker_offset\u001b[1;34m(str1, str2, start_date)\u001b[0m\n\u001b[0;32m     10\u001b[0m     \u001b[0ma\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0myf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdownload\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtickers\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mstr1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minterval\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mglobal_interval\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstart\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mstart_date\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mend\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mend_date\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mprogress\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     11\u001b[0m     \u001b[0mb\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0myf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mdownload\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mtickers\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mstr2\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0minterval\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mglobal_interval\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstart\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mstart_date\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mend\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mend_date\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mprogress\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mFalse\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 12\u001b[1;33m     \u001b[0mdf\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0ma\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mmerge\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mb\u001b[0m\u001b[1;33m,\u001b[0m\u001b[0mhow\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"inner\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mleft_on\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"Datetime\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mright_on\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;34m\"Datetime\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0msuffixes\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;34m\"_a\"\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m\"_b\"\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     13\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     14\u001b[0m     \u001b[1;31m## Use only the first half of the day\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\Users\\Andrew\\Documents\\git_repos\\BinancePublicData\\venv\\lib\\site-packages\\pandas\\core\\frame.py\u001b[0m in \u001b[0;36mmerge\u001b[1;34m(self, right, how, on, left_on, right_on, left_index, right_index, sort, suffixes, copy, indicator, validate)\u001b[0m\n\u001b[0;32m   9201\u001b[0m             \u001b[0mcopy\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   9202\u001b[0m             \u001b[0mindicator\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mindicator\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 9203\u001b[1;33m             \u001b[0mvalidate\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mvalidate\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   9204\u001b[0m         )\n\u001b[0;32m   9205\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\Users\\Andrew\\Documents\\git_repos\\BinancePublicData\\venv\\lib\\site-packages\\pandas\\core\\reshape\\merge.py\u001b[0m in \u001b[0;36mmerge\u001b[1;34m(left, right, how, on, left_on, right_on, left_index, right_index, sort, suffixes, copy, indicator, validate)\u001b[0m\n\u001b[0;32m    117\u001b[0m         \u001b[0mcopy\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mcopy\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    118\u001b[0m         \u001b[0mindicator\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mindicator\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 119\u001b[1;33m         \u001b[0mvalidate\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mvalidate\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    120\u001b[0m     )\n\u001b[0;32m    121\u001b[0m     \u001b[1;32mreturn\u001b[0m \u001b[0mop\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_result\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\Users\\Andrew\\Documents\\git_repos\\BinancePublicData\\venv\\lib\\site-packages\\pandas\\core\\reshape\\merge.py\u001b[0m in \u001b[0;36m__init__\u001b[1;34m(self, left, right, how, on, left_on, right_on, axis, left_index, right_index, sort, suffixes, copy, indicator, validate)\u001b[0m\n\u001b[0;32m    697\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mright_join_keys\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    698\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mjoin_names\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 699\u001b[1;33m         ) = self._get_merge_keys()\n\u001b[0m\u001b[0;32m    700\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    701\u001b[0m         \u001b[1;31m# validate the merge keys dtypes. We may need to coerce\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\Users\\Andrew\\Documents\\git_repos\\BinancePublicData\\venv\\lib\\site-packages\\pandas\\core\\reshape\\merge.py\u001b[0m in \u001b[0;36m_get_merge_keys\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m   1094\u001b[0m                     \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mis_rkey\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrk\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1095\u001b[0m                         \u001b[1;32mif\u001b[0m \u001b[0mrk\u001b[0m \u001b[1;32mis\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1096\u001b[1;33m                             \u001b[0mright_keys\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mright\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_get_label_or_level_values\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mrk\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1097\u001b[0m                         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1098\u001b[0m                             \u001b[1;31m# work-around for merge_asof(right_index=True)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mc:\\Users\\Andrew\\Documents\\git_repos\\BinancePublicData\\venv\\lib\\site-packages\\pandas\\core\\generic.py\u001b[0m in \u001b[0;36m_get_label_or_level_values\u001b[1;34m(self, key, axis)\u001b[0m\n\u001b[0;32m   1777\u001b[0m             \u001b[0mvalues\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0maxes\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0maxis\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mget_level_values\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_values\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1778\u001b[0m         \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m-> 1779\u001b[1;33m             \u001b[1;32mraise\u001b[0m \u001b[0mKeyError\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mkey\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m   1780\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m   1781\u001b[0m         \u001b[1;31m# Check for duplicates\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyError\u001b[0m: 'Datetime'"
     ]
    }
   ],
   "source": [
    "#Earliest is 30 days before current day\n",
    "starting_date = \"2022-04-16\"\n",
    "delt = timedelta(days=1)\n",
    "current_date = pd.Timestamp(starting_date)\n",
    "for i in range(17):\n",
    "    cur_date_str = str(current_date).split()[0]\n",
    "    if current_date.weekday() >= 5:\n",
    "        #Ignore Saturday and Sunday if QQQ is included\n",
    "        print(f\"Ignoring {cur_date_str}\")\n",
    "        current_date += delt\n",
    "        continue\n",
    "    history[cur_date_str] = backtest_one_day(cur_date_str)\n",
    "    print(\"Done with\",cur_date_str)\n",
    "    current_date += delt\n",
    "list(history.keys())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_most_profitable_tickers(n, symbs, pnls):\n",
    "    inds = np.argsort(-np.array(pnls))\n",
    "    best = []\n",
    "    for i in range(n):\n",
    "        ind = inds[i]\n",
    "        print(f\"{symbs[ind]} | {pnls[ind]:.2f}\")\n",
    "        best.append(symbs[ind])\n",
    "    return best\n",
    "\n",
    "def get_least_profitable_tickers(n, symbs, pnls):\n",
    "    inds = np.argsort(np.array(pnls))\n",
    "    best = []\n",
    "    for i in range(n):\n",
    "        ind = inds[i]\n",
    "        print(f\"{symbs[ind]} | {pnls[ind]:.2f}\")\n",
    "        best.append(symbs[ind])\n",
    "    return best"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Results\n",
    "\n",
    "Overall, the trades were mostly positive, winning 52% of the trades made from our aforementioned strategy. Our strategy outperformed BTC heavily during the past 30 days, returning +66% (assuming each trade used 100% of account) while BTC returned -16%. \n",
    "\n",
    "Profit factor was 1.1 which is good considering the state of the market overall and that our strategy only went long. We considered a number of indicators that could work and that led to a working strategy. There are additional optimizations we can do, like online updates to cross correlation as data is coming in. Then, we still would want to try out additional hyperparameter tuning and try to limit losses by placing limits on the slope loss for indicators that we consider to be valid.\n",
    "\n",
    "Moreover, given the downtrend in the current market, it would be better if our strategy could play both long and short. This would just mean adding an additional entry parameter which the inverse of the long entry. Looking at the results by day, we can see a clear positive correlation between BTC's % change that day and our strategy's profitability. Thus, we hypothesize that adding a short entry signal would eliminate this long bias and make our strategy much more profitable in the current market environment. \n",
    "\n",
    "Adding on to this, again looking at the results by day, we can see that there are clearly days when the strategy works well (winrate >60%, avg win >> avg loss on those days) and when the strategy performs poorly (40% or less winrate, avg loss >> avg win). In the future, it would be good to add some way to filter the signals so that trades that are less likely to be profitable can be ignored. \n",
    "\n",
    "However, when looking at trades by token, we can see that QQQ-crypto pairings did not generate any edge. The best performing pairings were BTC-alt or alt-alt.\n",
    "\n",
    "\n",
    "## Conclusion\n",
    "\n",
    "The results of this study were promising. We were able to generate a profit trading only long during a period of time when the markets were extremely bearish. Thus we can reasonably conclude that there does seem to be a significant edge in cross correlation trading. Our pilot study however was very brief and only scratched the surface. There is a lot more research that can and should be done on this topic, including examining different stocks and pairings, adding more and tuning existing hyperparameters, adding more features and indicators that could have predictive power to act as signals or filters to trigger/ignore trades, and more. \n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All Trades (past 30 days) | Total P/L % :: 66.01%\n",
      "Winrate 52.43% | 3878 Trades, 1847 Winners, 1676 Losers, 355 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.49%\n",
      "Winners | Avg Trade 0.34% | SD 0.38%\n",
      "Losers | Avg Trade -0.33% | SD 0.39%\n",
      "Profit Factor: 1.1190124798731227 | Gross Profit: 620.64% | Gross Loss: -554.63%\n",
      "----------\n",
      "----------\n"
     ]
    }
   ],
   "source": [
    "first_index = list(history.keys())[0]\n",
    "all_pairings = history[first_index][0]\n",
    "all_trades = []\n",
    "all_pnls = np.zeros(len(history[first_index][1]))\n",
    "for _, pnls, pnl_dict, trades, trades_dict in history.values():\n",
    "    for pair in trades_dict.keys():\n",
    "        if \"XRP-USD\" in pair or \"ADA-USD\" in pair:\n",
    "            continue\n",
    "        all_trades += trades_dict[pair]\n",
    "        all_pnls += np.array(pnl_dict[pair])\n",
    "\n",
    "all_trades = np.array(all_trades)\n",
    "stats(all_trades, \"All Trades (past 30 days)\")\n",
    "print(\"----------\")\n",
    "print(\"----------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2022-04-01 | Total P/L % :: 36.58%\n",
      "Winrate 70.06% | 177 Trades, 117 Winners, 50 Losers, 10 Breakeven\n",
      "All Trades | Avg Trade 0.21% | SD 0.46%\n",
      "Winners | Avg Trade 0.42% | SD 0.39%\n",
      "Losers | Avg Trade -0.25% | SD 0.24%\n",
      "Profit Factor: 3.9319857573115047 | Gross Profit: 49.05% | Gross Loss: -12.47%\n",
      "-------------------------\n",
      "2022-04-04 | Total P/L % :: 2.21%\n",
      "Winrate 46.84% | 214 Trades, 89 Winners, 101 Losers, 24 Breakeven\n",
      "All Trades | Avg Trade 0.01% | SD 0.31%\n",
      "Winners | Avg Trade 0.26% | SD 0.31%\n",
      "Losers | Avg Trade -0.20% | SD 0.14%\n",
      "Profit Factor: 1.107483179300907 | Gross Profit: 22.72% | Gross Loss: -20.52%\n",
      "-------------------------\n",
      "2022-04-05 | Total P/L % :: -12.40%\n",
      "Winrate 40.27% | 242 Trades, 91 Winners, 135 Losers, 16 Breakeven\n",
      "All Trades | Avg Trade -0.05% | SD 0.31%\n",
      "Winners | Avg Trade 0.19% | SD 0.17%\n",
      "Losers | Avg Trade -0.22% | SD 0.30%\n",
      "Profit Factor: 0.5766356079808392 | Gross Profit: 16.88% | Gross Loss: -29.28%\n",
      "-------------------------\n",
      "2022-04-06 | Total P/L % :: -12.78%\n",
      "Winrate 46.46% | 230 Trades, 92 Winners, 106 Losers, 32 Breakeven\n",
      "All Trades | Avg Trade -0.06% | SD 0.51%\n",
      "Winners | Avg Trade 0.38% | SD 0.37%\n",
      "Losers | Avg Trade -0.45% | SD 0.37%\n",
      "Profit Factor: 0.7301444979133608 | Gross Profit: 34.58% | Gross Loss: -47.36%\n",
      "-------------------------\n",
      "2022-04-07 | Total P/L % :: 21.72%\n",
      "Winrate 57.43% | 216 Trades, 116 Winners, 86 Losers, 14 Breakeven\n",
      "All Trades | Avg Trade 0.10% | SD 0.42%\n",
      "Winners | Avg Trade 0.39% | SD 0.32%\n",
      "Losers | Avg Trade -0.27% | SD 0.24%\n",
      "Profit Factor: 1.9458476419543542 | Gross Profit: 44.68% | Gross Loss: -22.96%\n",
      "-------------------------\n",
      "2022-04-08 | Total P/L % :: -6.75%\n",
      "Winrate 44.00% | 222 Trades, 88 Winners, 112 Losers, 22 Breakeven\n",
      "All Trades | Avg Trade -0.03% | SD 0.47%\n",
      "Winners | Avg Trade 0.38% | SD 0.31%\n",
      "Losers | Avg Trade -0.36% | SD 0.34%\n",
      "Profit Factor: 0.8329436632764522 | Gross Profit: 33.64% | Gross Loss: -40.39%\n",
      "-------------------------\n",
      "2022-04-11 | Total P/L % :: -17.54%\n",
      "Winrate 39.47% | 202 Trades, 75 Winners, 115 Losers, 12 Breakeven\n",
      "All Trades | Avg Trade -0.09% | SD 0.33%\n",
      "Winners | Avg Trade 0.21% | SD 0.16%\n",
      "Losers | Avg Trade -0.29% | SD 0.27%\n",
      "Profit Factor: 0.4786453794275542 | Gross Profit: 16.10% | Gross Loss: -33.64%\n",
      "-------------------------\n",
      "2022-04-12 | Total P/L % :: 47.18%\n",
      "Winrate 70.83% | 128 Trades, 85 Winners, 35 Losers, 8 Breakeven\n",
      "All Trades | Avg Trade 0.37% | SD 0.72%\n",
      "Winners | Avg Trade 0.69% | SD 0.65%\n",
      "Losers | Avg Trade -0.33% | SD 0.27%\n",
      "Profit Factor: 5.027999871126033 | Gross Profit: 58.89% | Gross Loss: -11.71%\n",
      "-------------------------\n",
      "2022-04-13 | Total P/L % :: 19.47%\n",
      "Winrate 59.55% | 236 Trades, 131 Winners, 89 Losers, 16 Breakeven\n",
      "All Trades | Avg Trade 0.08% | SD 0.30%\n",
      "Winners | Avg Trade 0.29% | SD 0.21%\n",
      "Losers | Avg Trade -0.20% | SD 0.19%\n",
      "Profit Factor: 2.074185345093431 | Gross Profit: 37.60% | Gross Loss: -18.13%\n",
      "-------------------------\n",
      "2022-04-14 | Total P/L % :: -12.96%\n",
      "Winrate 43.40% | 232 Trades, 92 Winners, 120 Losers, 20 Breakeven\n",
      "All Trades | Avg Trade -0.06% | SD 0.35%\n",
      "Winners | Avg Trade 0.24% | SD 0.23%\n",
      "Losers | Avg Trade -0.29% | SD 0.26%\n",
      "Profit Factor: 0.6257156718884684 | Gross Profit: 21.67% | Gross Loss: -34.63%\n",
      "-------------------------\n",
      "2022-04-18 | Total P/L % :: -13.45%\n",
      "Winrate 48.44% | 141 Trades, 62 Winners, 66 Losers, 13 Breakeven\n",
      "All Trades | Avg Trade -0.10% | SD 0.48%\n",
      "Winners | Avg Trade 0.24% | SD 0.20%\n",
      "Losers | Avg Trade -0.43% | SD 0.48%\n",
      "Profit Factor: 0.5294304496825812 | Gross Profit: 15.13% | Gross Loss: -28.58%\n",
      "-------------------------\n",
      "2022-04-19 | Total P/L % :: 6.03%\n",
      "Winrate 51.83% | 240 Trades, 113 Winners, 105 Losers, 22 Breakeven\n",
      "All Trades | Avg Trade 0.03% | SD 0.30%\n",
      "Winners | Avg Trade 0.24% | SD 0.26%\n",
      "Losers | Avg Trade -0.20% | SD 0.19%\n",
      "Profit Factor: 1.291437473578219 | Gross Profit: 26.71% | Gross Loss: -20.68%\n",
      "-------------------------\n",
      "2022-04-20 | Total P/L % :: 3.90%\n",
      "Winrate 60.44% | 202 Trades, 110 Winners, 72 Losers, 20 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.26%\n",
      "Winners | Avg Trade 0.18% | SD 0.21%\n",
      "Losers | Avg Trade -0.22% | SD 0.18%\n",
      "Profit Factor: 1.2475190836606 | Gross Profit: 19.67% | Gross Loss: -15.77%\n",
      "-------------------------\n",
      "2022-04-21 | Total P/L % :: -3.19%\n",
      "Winrate 47.45% | 214 Trades, 93 Winners, 103 Losers, 18 Breakeven\n",
      "All Trades | Avg Trade -0.01% | SD 0.39%\n",
      "Winners | Avg Trade 0.30% | SD 0.26%\n",
      "Losers | Avg Trade -0.30% | SD 0.28%\n",
      "Profit Factor: 0.8972416351266185 | Gross Profit: 27.83% | Gross Loss: -31.02%\n",
      "-------------------------\n",
      "2022-04-22 | Total P/L % :: -19.15%\n",
      "Winrate 60.71% | 78 Trades, 34 Winners, 22 Losers, 22 Breakeven\n",
      "All Trades | Avg Trade -0.25% | SD 0.87%\n",
      "Winners | Avg Trade 0.26% | SD 0.18%\n",
      "Losers | Avg Trade -1.27% | SD 1.06%\n",
      "Profit Factor: 0.31487688849832135 | Gross Profit: 8.80% | Gross Loss: -27.95%\n",
      "-------------------------\n",
      "2022-04-25 | Total P/L % :: -57.14%\n",
      "Winrate 25.42% | 138 Trades, 30 Winners, 88 Losers, 20 Breakeven\n",
      "All Trades | Avg Trade -0.41% | SD 0.70%\n",
      "Winners | Avg Trade 0.18% | SD 0.24%\n",
      "Losers | Avg Trade -0.71% | SD 0.71%\n",
      "Profit Factor: 0.08545258759985362 | Gross Profit: 5.34% | Gross Loss: -62.48%\n",
      "-------------------------\n",
      "2022-04-26 | Total P/L % :: 29.09%\n",
      "Winrate 60.42% | 208 Trades, 116 Winners, 76 Losers, 16 Breakeven\n",
      "All Trades | Avg Trade 0.14% | SD 0.62%\n",
      "Winners | Avg Trade 0.40% | SD 0.70%\n",
      "Losers | Avg Trade -0.23% | SD 0.24%\n",
      "Profit Factor: 2.6674711504111763 | Gross Profit: 46.54% | Gross Loss: -17.45%\n",
      "-------------------------\n",
      "2022-04-27 | Total P/L % :: 36.95%\n",
      "Winrate 70.39% | 170 Trades, 107 Winners, 45 Losers, 18 Breakeven\n",
      "All Trades | Avg Trade 0.22% | SD 0.50%\n",
      "Winners | Avg Trade 0.47% | SD 0.41%\n",
      "Losers | Avg Trade -0.31% | SD 0.30%\n",
      "Profit Factor: 3.6698424122341247 | Gross Profit: 50.79% | Gross Loss: -13.84%\n",
      "-------------------------\n",
      "2022-04-28 | Total P/L % :: 32.76%\n",
      "Winrate 66.89% | 166 Trades, 99 Winners, 49 Losers, 18 Breakeven\n",
      "All Trades | Avg Trade 0.20% | SD 0.57%\n",
      "Winners | Avg Trade 0.49% | SD 0.54%\n",
      "Losers | Avg Trade -0.32% | SD 0.26%\n",
      "Profit Factor: 3.1162130004983775 | Gross Profit: 48.23% | Gross Loss: -15.48%\n",
      "-------------------------\n",
      "2022-04-29 | Total P/L % :: -14.52%\n",
      "Winrate 51.44% | 222 Trades, 107 Winners, 101 Losers, 14 Breakeven\n",
      "All Trades | Avg Trade -0.07% | SD 0.55%\n",
      "Winners | Avg Trade 0.33% | SD 0.24%\n",
      "Losers | Avg Trade -0.50% | SD 0.51%\n",
      "Profit Factor: 0.7112476507399474 | Gross Profit: 35.77% | Gross Loss: -50.29%\n",
      "-------------------------\n"
     ]
    }
   ],
   "source": [
    "for day in history.keys():\n",
    "    _, pnls, pnls_dict, trades, trades_dict = history[day]\n",
    "    trades_that_day = []\n",
    "    for pair in trades_dict.keys():\n",
    "        if \"XRP-USD\" in pair or \"ADA-USD\" in pair:\n",
    "            continue\n",
    "        trades_that_day += trades_dict[pair]\n",
    "    stats(np.array(trades_that_day), day)\n",
    "    print(\"-------------------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\n2022-04-01 | Total P/L % :: 36.58%\\nWinrate 70.06% | 177 Trades, 117 Winners, 50 Losers, 10 Breakeven\\nAll Trades | Avg Trade 0.21% | SD 0.46%\\nWinners | Avg Trade 0.42% | SD 0.39%\\nLosers | Avg Trade -0.25% | SD 0.24%\\nProfit Factor: 3.9319857573115047 | Gross Profit: 49.05% | Gross Loss: -12.47%\\n-------------------------\\n2022-04-04 | Total P/L % :: 2.21%\\nWinrate 46.84% | 214 Trades, 89 Winners, 101 Losers, 24 Breakeven\\nAll Trades | Avg Trade 0.01% | SD 0.31%\\nWinners | Avg Trade 0.26% | SD 0.31%\\nLosers | Avg Trade -0.20% | SD 0.14%\\nProfit Factor: 1.107483179300907 | Gross Profit: 22.72% | Gross Loss: -20.52%\\n-------------------------\\n2022-04-05 | Total P/L % :: -12.40%\\nWinrate 40.27% | 242 Trades, 91 Winners, 135 Losers, 16 Breakeven\\nAll Trades | Avg Trade -0.05% | SD 0.31%\\nWinners | Avg Trade 0.19% | SD 0.17%\\nLosers | Avg Trade -0.22% | SD 0.30%\\nProfit Factor: 0.5766356079808392 | Gross Profit: 16.88% | Gross Loss: -29.28%\\n-------------------------\\n2022-04-06 | Total P/L % :: -12.78%\\nWinrate 46.46% | 230 Trades, 92 Winners, 106 Losers, 32 Breakeven\\nAll Trades | Avg Trade -0.06% | SD 0.51%\\nWinners | Avg Trade 0.38% | SD 0.37%\\nLosers | Avg Trade -0.45% | SD 0.37%\\nProfit Factor: 0.7301444979133608 | Gross Profit: 34.58% | Gross Loss: -47.36%\\n-------------------------\\n2022-04-07 | Total P/L % :: 21.72%\\nWinrate 57.43% | 216 Trades, 116 Winners, 86 Losers, 14 Breakeven\\nAll Trades | Avg Trade 0.10% | SD 0.42%\\nWinners | Avg Trade 0.39% | SD 0.32%\\nLosers | Avg Trade -0.27% | SD 0.24%\\nProfit Factor: 1.9458476419543542 | Gross Profit: 44.68% | Gross Loss: -22.96%\\n-------------------------\\n2022-04-08 | Total P/L % :: -6.75%\\nWinrate 44.00% | 222 Trades, 88 Winners, 112 Losers, 22 Breakeven\\nAll Trades | Avg Trade -0.03% | SD 0.47%\\nWinners | Avg Trade 0.38% | SD 0.31%\\nLosers | Avg Trade -0.36% | SD 0.34%\\nProfit Factor: 0.8329436632764522 | Gross Profit: 33.64% | Gross Loss: -40.39%\\n-------------------------\\n2022-04-11 | Total P/L % :: -17.54%\\nWinrate 39.47% | 202 Trades, 75 Winners, 115 Losers, 12 Breakeven\\nAll Trades | Avg Trade -0.09% | SD 0.33%\\nWinners | Avg Trade 0.21% | SD 0.16%\\nLosers | Avg Trade -0.29% | SD 0.27%\\nProfit Factor: 0.4786453794275542 | Gross Profit: 16.10% | Gross Loss: -33.64%\\n-------------------------\\n2022-04-12 | Total P/L % :: 47.18%\\nWinrate 70.83% | 128 Trades, 85 Winners, 35 Losers, 8 Breakeven\\nAll Trades | Avg Trade 0.37% | SD 0.72%\\nWinners | Avg Trade 0.69% | SD 0.65%\\nLosers | Avg Trade -0.33% | SD 0.27%\\nProfit Factor: 5.027999871126033 | Gross Profit: 58.89% | Gross Loss: -11.71%\\n-------------------------\\n2022-04-13 | Total P/L % :: 19.47%\\nWinrate 59.55% | 236 Trades, 131 Winners, 89 Losers, 16 Breakeven\\nAll Trades | Avg Trade 0.08% | SD 0.30%\\nWinners | Avg Trade 0.29% | SD 0.21%\\nLosers | Avg Trade -0.20% | SD 0.19%\\nProfit Factor: 2.074185345093431 | Gross Profit: 37.60% | Gross Loss: -18.13%\\n-------------------------\\n2022-04-14 | Total P/L % :: -12.96%\\nWinrate 43.40% | 232 Trades, 92 Winners, 120 Losers, 20 Breakeven\\nAll Trades | Avg Trade -0.06% | SD 0.35%\\nWinners | Avg Trade 0.24% | SD 0.23%\\nLosers | Avg Trade -0.29% | SD 0.26%\\nProfit Factor: 0.6257156718884684 | Gross Profit: 21.67% | Gross Loss: -34.63%\\n-------------------------\\n2022-04-18 | Total P/L % :: -13.45%\\nWinrate 48.44% | 141 Trades, 62 Winners, 66 Losers, 13 Breakeven\\nAll Trades | Avg Trade -0.10% | SD 0.48%\\nWinners | Avg Trade 0.24% | SD 0.20%\\nLosers | Avg Trade -0.43% | SD 0.48%\\nProfit Factor: 0.5294304496825812 | Gross Profit: 15.13% | Gross Loss: -28.58%\\n-------------------------\\n2022-04-19 | Total P/L % :: 6.03%\\nWinrate 51.83% | 240 Trades, 113 Winners, 105 Losers, 22 Breakeven\\nAll Trades | Avg Trade 0.03% | SD 0.30%\\nWinners | Avg Trade 0.24% | SD 0.26%\\nLosers | Avg Trade -0.20% | SD 0.19%\\nProfit Factor: 1.291437473578219 | Gross Profit: 26.71% | Gross Loss: -20.68%\\n-------------------------\\n2022-04-20 | Total P/L % :: 3.90%\\nWinrate 60.44% | 202 Trades, 110 Winners, 72 Losers, 20 Breakeven\\nAll Trades | Avg Trade 0.02% | SD 0.26%\\nWinners | Avg Trade 0.18% | SD 0.21%\\nLosers | Avg Trade -0.22% | SD 0.18%\\nProfit Factor: 1.2475190836606 | Gross Profit: 19.67% | Gross Loss: -15.77%\\n-------------------------\\n2022-04-21 | Total P/L % :: -3.19%\\nWinrate 47.45% | 214 Trades, 93 Winners, 103 Losers, 18 Breakeven\\nAll Trades | Avg Trade -0.01% | SD 0.39%\\nWinners | Avg Trade 0.30% | SD 0.26%\\nLosers | Avg Trade -0.30% | SD 0.28%\\nProfit Factor: 0.8972416351266185 | Gross Profit: 27.83% | Gross Loss: -31.02%\\n-------------------------\\n2022-04-22 | Total P/L % :: -19.15%\\nWinrate 60.71% | 78 Trades, 34 Winners, 22 Losers, 22 Breakeven\\nAll Trades | Avg Trade -0.25% | SD 0.87%\\nWinners | Avg Trade 0.26% | SD 0.18%\\nLosers | Avg Trade -1.27% | SD 1.06%\\nProfit Factor: 0.31487688849832135 | Gross Profit: 8.80% | Gross Loss: -27.95%\\n-------------------------\\n2022-04-25 | Total P/L % :: -57.14%\\nWinrate 25.42% | 138 Trades, 30 Winners, 88 Losers, 20 Breakeven\\nAll Trades | Avg Trade -0.41% | SD 0.70%\\nWinners | Avg Trade 0.18% | SD 0.24%\\nLosers | Avg Trade -0.71% | SD 0.71%\\nProfit Factor: 0.08545258759985362 | Gross Profit: 5.34% | Gross Loss: -62.48%\\n-------------------------\\n2022-04-26 | Total P/L % :: 29.09%\\nWinrate 60.42% | 208 Trades, 116 Winners, 76 Losers, 16 Breakeven\\nAll Trades | Avg Trade 0.14% | SD 0.62%\\nWinners | Avg Trade 0.40% | SD 0.70%\\nLosers | Avg Trade -0.23% | SD 0.24%\\nProfit Factor: 2.6674711504111763 | Gross Profit: 46.54% | Gross Loss: -17.45%\\n-------------------------\\n2022-04-27 | Total P/L % :: 36.95%\\nWinrate 70.39% | 170 Trades, 107 Winners, 45 Losers, 18 Breakeven\\nAll Trades | Avg Trade 0.22% | SD 0.50%\\nWinners | Avg Trade 0.47% | SD 0.41%\\nLosers | Avg Trade -0.31% | SD 0.30%\\nProfit Factor: 3.6698424122341247 | Gross Profit: 50.79% | Gross Loss: -13.84%\\n-------------------------\\n2022-04-28 | Total P/L % :: 32.76%\\nWinrate 66.89% | 166 Trades, 99 Winners, 49 Losers, 18 Breakeven\\nAll Trades | Avg Trade 0.20% | SD 0.57%\\nWinners | Avg Trade 0.49% | SD 0.54%\\nLosers | Avg Trade -0.32% | SD 0.26%\\nProfit Factor: 3.1162130004983775 | Gross Profit: 48.23% | Gross Loss: -15.48%\\n-------------------------\\n2022-04-29 | Total P/L % :: -14.52%\\nWinrate 51.44% | 222 Trades, 107 Winners, 101 Losers, 14 Breakeven\\nAll Trades | Avg Trade -0.07% | SD 0.55%\\nWinners | Avg Trade 0.33% | SD 0.24%\\nLosers | Avg Trade -0.50% | SD 0.51%\\nProfit Factor: 0.7112476507399474 | Gross Profit: 35.77% | Gross Loss: -50.29%\\n-------------------------\\n\\n'"
      ]
     },
     "execution_count": 181,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "2022-04-01 | Total P/L % :: 36.58%\n",
    "Winrate 70.06% | 177 Trades, 117 Winners, 50 Losers, 10 Breakeven\n",
    "All Trades | Avg Trade 0.21% | SD 0.46%\n",
    "Winners | Avg Trade 0.42% | SD 0.39%\n",
    "Losers | Avg Trade -0.25% | SD 0.24%\n",
    "Profit Factor: 3.9319857573115047 | Gross Profit: 49.05% | Gross Loss: -12.47%\n",
    "-------------------------\n",
    "2022-04-04 | Total P/L % :: 2.21%\n",
    "Winrate 46.84% | 214 Trades, 89 Winners, 101 Losers, 24 Breakeven\n",
    "All Trades | Avg Trade 0.01% | SD 0.31%\n",
    "Winners | Avg Trade 0.26% | SD 0.31%\n",
    "Losers | Avg Trade -0.20% | SD 0.14%\n",
    "Profit Factor: 1.107483179300907 | Gross Profit: 22.72% | Gross Loss: -20.52%\n",
    "-------------------------\n",
    "2022-04-05 | Total P/L % :: -12.40%\n",
    "Winrate 40.27% | 242 Trades, 91 Winners, 135 Losers, 16 Breakeven\n",
    "All Trades | Avg Trade -0.05% | SD 0.31%\n",
    "Winners | Avg Trade 0.19% | SD 0.17%\n",
    "Losers | Avg Trade -0.22% | SD 0.30%\n",
    "Profit Factor: 0.5766356079808392 | Gross Profit: 16.88% | Gross Loss: -29.28%\n",
    "-------------------------\n",
    "2022-04-06 | Total P/L % :: -12.78%\n",
    "Winrate 46.46% | 230 Trades, 92 Winners, 106 Losers, 32 Breakeven\n",
    "All Trades | Avg Trade -0.06% | SD 0.51%\n",
    "Winners | Avg Trade 0.38% | SD 0.37%\n",
    "Losers | Avg Trade -0.45% | SD 0.37%\n",
    "Profit Factor: 0.7301444979133608 | Gross Profit: 34.58% | Gross Loss: -47.36%\n",
    "-------------------------\n",
    "2022-04-07 | Total P/L % :: 21.72%\n",
    "Winrate 57.43% | 216 Trades, 116 Winners, 86 Losers, 14 Breakeven\n",
    "All Trades | Avg Trade 0.10% | SD 0.42%\n",
    "Winners | Avg Trade 0.39% | SD 0.32%\n",
    "Losers | Avg Trade -0.27% | SD 0.24%\n",
    "Profit Factor: 1.9458476419543542 | Gross Profit: 44.68% | Gross Loss: -22.96%\n",
    "-------------------------\n",
    "2022-04-08 | Total P/L % :: -6.75%\n",
    "Winrate 44.00% | 222 Trades, 88 Winners, 112 Losers, 22 Breakeven\n",
    "All Trades | Avg Trade -0.03% | SD 0.47%\n",
    "Winners | Avg Trade 0.38% | SD 0.31%\n",
    "Losers | Avg Trade -0.36% | SD 0.34%\n",
    "Profit Factor: 0.8329436632764522 | Gross Profit: 33.64% | Gross Loss: -40.39%\n",
    "-------------------------\n",
    "2022-04-11 | Total P/L % :: -17.54%\n",
    "Winrate 39.47% | 202 Trades, 75 Winners, 115 Losers, 12 Breakeven\n",
    "All Trades | Avg Trade -0.09% | SD 0.33%\n",
    "Winners | Avg Trade 0.21% | SD 0.16%\n",
    "Losers | Avg Trade -0.29% | SD 0.27%\n",
    "Profit Factor: 0.4786453794275542 | Gross Profit: 16.10% | Gross Loss: -33.64%\n",
    "-------------------------\n",
    "2022-04-12 | Total P/L % :: 47.18%\n",
    "Winrate 70.83% | 128 Trades, 85 Winners, 35 Losers, 8 Breakeven\n",
    "All Trades | Avg Trade 0.37% | SD 0.72%\n",
    "Winners | Avg Trade 0.69% | SD 0.65%\n",
    "Losers | Avg Trade -0.33% | SD 0.27%\n",
    "Profit Factor: 5.027999871126033 | Gross Profit: 58.89% | Gross Loss: -11.71%\n",
    "-------------------------\n",
    "2022-04-13 | Total P/L % :: 19.47%\n",
    "Winrate 59.55% | 236 Trades, 131 Winners, 89 Losers, 16 Breakeven\n",
    "All Trades | Avg Trade 0.08% | SD 0.30%\n",
    "Winners | Avg Trade 0.29% | SD 0.21%\n",
    "Losers | Avg Trade -0.20% | SD 0.19%\n",
    "Profit Factor: 2.074185345093431 | Gross Profit: 37.60% | Gross Loss: -18.13%\n",
    "-------------------------\n",
    "2022-04-14 | Total P/L % :: -12.96%\n",
    "Winrate 43.40% | 232 Trades, 92 Winners, 120 Losers, 20 Breakeven\n",
    "All Trades | Avg Trade -0.06% | SD 0.35%\n",
    "Winners | Avg Trade 0.24% | SD 0.23%\n",
    "Losers | Avg Trade -0.29% | SD 0.26%\n",
    "Profit Factor: 0.6257156718884684 | Gross Profit: 21.67% | Gross Loss: -34.63%\n",
    "-------------------------\n",
    "2022-04-18 | Total P/L % :: -13.45%\n",
    "Winrate 48.44% | 141 Trades, 62 Winners, 66 Losers, 13 Breakeven\n",
    "All Trades | Avg Trade -0.10% | SD 0.48%\n",
    "Winners | Avg Trade 0.24% | SD 0.20%\n",
    "Losers | Avg Trade -0.43% | SD 0.48%\n",
    "Profit Factor: 0.5294304496825812 | Gross Profit: 15.13% | Gross Loss: -28.58%\n",
    "-------------------------\n",
    "2022-04-19 | Total P/L % :: 6.03%\n",
    "Winrate 51.83% | 240 Trades, 113 Winners, 105 Losers, 22 Breakeven\n",
    "All Trades | Avg Trade 0.03% | SD 0.30%\n",
    "Winners | Avg Trade 0.24% | SD 0.26%\n",
    "Losers | Avg Trade -0.20% | SD 0.19%\n",
    "Profit Factor: 1.291437473578219 | Gross Profit: 26.71% | Gross Loss: -20.68%\n",
    "-------------------------\n",
    "2022-04-20 | Total P/L % :: 3.90%\n",
    "Winrate 60.44% | 202 Trades, 110 Winners, 72 Losers, 20 Breakeven\n",
    "All Trades | Avg Trade 0.02% | SD 0.26%\n",
    "Winners | Avg Trade 0.18% | SD 0.21%\n",
    "Losers | Avg Trade -0.22% | SD 0.18%\n",
    "Profit Factor: 1.2475190836606 | Gross Profit: 19.67% | Gross Loss: -15.77%\n",
    "-------------------------\n",
    "2022-04-21 | Total P/L % :: -3.19%\n",
    "Winrate 47.45% | 214 Trades, 93 Winners, 103 Losers, 18 Breakeven\n",
    "All Trades | Avg Trade -0.01% | SD 0.39%\n",
    "Winners | Avg Trade 0.30% | SD 0.26%\n",
    "Losers | Avg Trade -0.30% | SD 0.28%\n",
    "Profit Factor: 0.8972416351266185 | Gross Profit: 27.83% | Gross Loss: -31.02%\n",
    "-------------------------\n",
    "2022-04-22 | Total P/L % :: -19.15%\n",
    "Winrate 60.71% | 78 Trades, 34 Winners, 22 Losers, 22 Breakeven\n",
    "All Trades | Avg Trade -0.25% | SD 0.87%\n",
    "Winners | Avg Trade 0.26% | SD 0.18%\n",
    "Losers | Avg Trade -1.27% | SD 1.06%\n",
    "Profit Factor: 0.31487688849832135 | Gross Profit: 8.80% | Gross Loss: -27.95%\n",
    "-------------------------\n",
    "2022-04-25 | Total P/L % :: -57.14%\n",
    "Winrate 25.42% | 138 Trades, 30 Winners, 88 Losers, 20 Breakeven\n",
    "All Trades | Avg Trade -0.41% | SD 0.70%\n",
    "Winners | Avg Trade 0.18% | SD 0.24%\n",
    "Losers | Avg Trade -0.71% | SD 0.71%\n",
    "Profit Factor: 0.08545258759985362 | Gross Profit: 5.34% | Gross Loss: -62.48%\n",
    "-------------------------\n",
    "2022-04-26 | Total P/L % :: 29.09%\n",
    "Winrate 60.42% | 208 Trades, 116 Winners, 76 Losers, 16 Breakeven\n",
    "All Trades | Avg Trade 0.14% | SD 0.62%\n",
    "Winners | Avg Trade 0.40% | SD 0.70%\n",
    "Losers | Avg Trade -0.23% | SD 0.24%\n",
    "Profit Factor: 2.6674711504111763 | Gross Profit: 46.54% | Gross Loss: -17.45%\n",
    "-------------------------\n",
    "2022-04-27 | Total P/L % :: 36.95%\n",
    "Winrate 70.39% | 170 Trades, 107 Winners, 45 Losers, 18 Breakeven\n",
    "All Trades | Avg Trade 0.22% | SD 0.50%\n",
    "Winners | Avg Trade 0.47% | SD 0.41%\n",
    "Losers | Avg Trade -0.31% | SD 0.30%\n",
    "Profit Factor: 3.6698424122341247 | Gross Profit: 50.79% | Gross Loss: -13.84%\n",
    "-------------------------\n",
    "2022-04-28 | Total P/L % :: 32.76%\n",
    "Winrate 66.89% | 166 Trades, 99 Winners, 49 Losers, 18 Breakeven\n",
    "All Trades | Avg Trade 0.20% | SD 0.57%\n",
    "Winners | Avg Trade 0.49% | SD 0.54%\n",
    "Losers | Avg Trade -0.32% | SD 0.26%\n",
    "Profit Factor: 3.1162130004983775 | Gross Profit: 48.23% | Gross Loss: -15.48%\n",
    "-------------------------\n",
    "2022-04-29 | Total P/L % :: -14.52%\n",
    "Winrate 51.44% | 222 Trades, 107 Winners, 101 Losers, 14 Breakeven\n",
    "All Trades | Avg Trade -0.07% | SD 0.55%\n",
    "Winners | Avg Trade 0.33% | SD 0.24%\n",
    "Losers | Avg Trade -0.50% | SD 0.51%\n",
    "Profit Factor: 0.7112476507399474 | Gross Profit: 35.77% | Gross Loss: -50.29%\n",
    "-------------------------\n",
    "\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BTC-USDDOGE-USD | 7.37\n",
      "HEX-USDBNB-USD | 7.15\n",
      "DOGE-USDBNB-USD | 6.23\n",
      "LUNA1-USDHEX-USD | 6.04\n",
      "AVAX-USDADA-USD | 5.85\n",
      "HEX-USDLUNA1-USD | 5.62\n",
      "LUNA1-USDBNB-USD | 5.05\n",
      "ADA-USDAVAX-USD | 4.95\n",
      "BTC-USDBNB-USD | 4.89\n",
      "BTC-USDALGO-USD | 4.88\n",
      "XRP-USDDOGE-USD | 4.66\n",
      "ETH-USDBNB-USD | 4.43\n",
      "BNB-USDLUNA1-USD | 4.27\n",
      "LUNA1-USDAVAX-USD | 3.96\n",
      "AVAX-USDHEX-USD | 3.86\n"
     ]
    }
   ],
   "source": [
    "best_pairs = get_most_profitable_tickers(15, all_pairings, all_pnls)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ALGO-USDAVAX-USD | -8.45\n",
      "ADA-USDDOGE-USD | -7.23\n",
      "ALGO-USDXRP-USD | -7.17\n",
      "BTC-USDXRP-USD | -6.93\n",
      "DOGE-USDXRP-USD | -6.79\n",
      "ADA-USDXRP-USD | -6.67\n",
      "HEX-USDXRP-USD | -5.33\n",
      "ALGO-USDADA-USD | -5.20\n",
      "BNB-USDAVAX-USD | -5.08\n",
      "BNB-USDADA-USD | -5.04\n",
      "LUNA1-USDXRP-USD | -5.00\n",
      "DOGE-USDADA-USD | -4.88\n",
      "XRP-USDADA-USD | -4.65\n",
      "AVAX-USDALGO-USD | -4.49\n",
      "ETH-USDBTC-USD | -4.17\n"
     ]
    }
   ],
   "source": [
    "worst_pairs = get_least_profitable_tickers(15, all_pairings, all_pnls)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Our trading strategy is simply comparing ith and (i+1)th slope, if the upwards and downwards diff is greater than a certain threshold, then we buy/sell y2 at the (i+1 timestamp + offset). Given we have already shifted the timestamp and joined both crypto dataframes, I just use y2's y2_original[diffs[i+1][1]] for buy and sell. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### We paid extra attention to data leackage, i.e. not using future knowledge to trade the current stock\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['QQQ',\n",
       " 'BTC-USD',\n",
       " 'ETH-USD',\n",
       " 'BNB-USD',\n",
       " 'XRP-USD',\n",
       " 'LUNA1-USD',\n",
       " 'ADA-USD',\n",
       " 'ALGO-USD',\n",
       " 'HEX-USD',\n",
       " 'AVAX-USD',\n",
       " 'DOGE-USD']"
      ]
     },
     "execution_count": 162,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokens = [\"QQQ\", \"BTC-USD\", \"ETH-USD\", \"BNB-USD\", \"XRP-USD\", \"LUNA1-USD\", \"ADA-USD\", \"ALGO-USD\", \"HEX-USD\", \"AVAX-USD\", \"DOGE-USD\"] # \n",
    "trades_by_token = {}\n",
    "for token in tokens:\n",
    "    for day in history.keys():\n",
    "        _, pnls, pnls_dict, trades, trades_dict = history[day]\n",
    "        for key in trades_dict.keys():\n",
    "            if token in key:\n",
    "                trades_by_token[token] = trades_by_token.get(token, []) + trades_dict[key]\n",
    "list(trades_by_token.keys())\n",
    "\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "QQQ | Total P/L % :: -4.67%\n",
      "Winrate 50.00% | 257 Trades, 71 Winners, 71 Losers, 115 Breakeven\n",
      "All Trades | Avg Trade -0.02% | SD 0.37%\n",
      "Winners | Avg Trade 0.36% | SD 0.24%\n",
      "Losers | Avg Trade -0.42% | SD 0.35%\n",
      "Profit Factor: 0.8439297984360714 | Gross Profit: 25.26% | Gross Loss: -29.93%\n",
      "----------------\n",
      "BTC-USD | Total P/L % :: 5.92%\n",
      "Winrate 52.60% | 1016 Trades, 485 Winners, 437 Losers, 94 Breakeven\n",
      "All Trades | Avg Trade 0.01% | SD 0.52%\n",
      "Winners | Avg Trade 0.30% | SD 0.40%\n",
      "Losers | Avg Trade -0.32% | SD 0.50%\n",
      "Profit Factor: 1.0425461208948326 | Gross Profit: 144.96% | Gross Loss: -139.04%\n",
      "----------------\n",
      "ETH-USD | Total P/L % :: 7.28%\n",
      "Winrate 49.40% | 1166 Trades, 536 Winners, 549 Losers, 81 Breakeven\n",
      "All Trades | Avg Trade 0.01% | SD 0.41%\n",
      "Winners | Avg Trade 0.28% | SD 0.33%\n",
      "Losers | Avg Trade -0.26% | SD 0.33%\n",
      "Profit Factor: 1.0501686433778488 | Gross Profit: 152.35% | Gross Loss: -145.07%\n",
      "----------------\n",
      "BNB-USD | Total P/L % :: 30.02%\n",
      "Winrate 53.90% | 1245 Trades, 629 Winners, 538 Losers, 78 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.44%\n",
      "Winners | Avg Trade 0.29% | SD 0.37%\n",
      "Losers | Avg Trade -0.29% | SD 0.33%\n",
      "Profit Factor: 1.1945617990587647 | Gross Profit: 184.34% | Gross Loss: -154.32%\n",
      "----------------\n",
      "LUNA1-USD | Total P/L % :: 50.68%\n",
      "Winrate 54.18% | 2358 Trades, 1180 Winners, 998 Losers, 180 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.49%\n",
      "Winners | Avg Trade 0.35% | SD 0.34%\n",
      "Losers | Avg Trade -0.36% | SD 0.39%\n",
      "Profit Factor: 1.1393385435967547 | Gross Profit: 414.38% | Gross Loss: -363.70%\n",
      "----------------\n",
      "ALGO-USD | Total P/L % :: -9.09%\n",
      "Winrate 51.15% | 1242 Trades, 577 Winners, 551 Losers, 114 Breakeven\n",
      "All Trades | Avg Trade -0.01% | SD 0.55%\n",
      "Winners | Avg Trade 0.34% | SD 0.36%\n",
      "Losers | Avg Trade -0.38% | SD 0.52%\n",
      "Profit Factor: 0.9563075557946746 | Gross Profit: 198.91% | Gross Loss: -208.00%\n",
      "----------------\n",
      "HEX-USD | Total P/L % :: 33.37%\n",
      "Winrate 52.29% | 1392 Trades, 684 Winners, 624 Losers, 84 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.44%\n",
      "Winners | Avg Trade 0.31% | SD 0.34%\n",
      "Losers | Avg Trade -0.29% | SD 0.32%\n",
      "Profit Factor: 1.184087073294854 | Gross Profit: 214.64% | Gross Loss: -181.27%\n",
      "----------------\n",
      "AVAX-USD | Total P/L % :: 6.22%\n",
      "Winrate 52.57% | 1235 Trades, 593 Winners, 535 Losers, 107 Breakeven\n",
      "All Trades | Avg Trade 0.01% | SD 0.49%\n",
      "Winners | Avg Trade 0.34% | SD 0.31%\n",
      "Losers | Avg Trade -0.37% | SD 0.43%\n",
      "Profit Factor: 1.0316654781994263 | Gross Profit: 202.75% | Gross Loss: -196.53%\n",
      "----------------\n",
      "DOGE-USD | Total P/L % :: -1.66%\n",
      "Winrate 49.95% | 1125 Trades, 514 Winners, 515 Losers, 96 Breakeven\n",
      "All Trades | Avg Trade -0.00% | SD 0.63%\n",
      "Winners | Avg Trade 0.41% | SD 0.51%\n",
      "Losers | Avg Trade -0.42% | SD 0.51%\n",
      "Profit Factor: 0.9922566583758843 | Gross Profit: 212.92% | Gross Loss: -214.58%\n",
      "----------------\n"
     ]
    }
   ],
   "source": [
    "for token in [\"QQQ\", \"BTC-USD\", \"ETH-USD\", \"BNB-USD\", \"LUNA1-USD\", \"ALGO-USD\", \"HEX-USD\", \"AVAX-USD\", \"DOGE-USD\"]:\n",
    "    stats(np.array(trades_by_token[token]), token)\n",
    "    print(\"----------------\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 184,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'\\nQQQ | Total P/L % :: -4.67%\\nWinrate 50.00% | 257 Trades, 71 Winners, 71 Losers, 115 Breakeven\\nAll Trades | Avg Trade -0.02% | SD 0.37%\\nWinners | Avg Trade 0.36% | SD 0.24%\\nLosers | Avg Trade -0.42% | SD 0.35%\\nProfit Factor: 0.8439297984360714 | Gross Profit: 25.26% | Gross Loss: -29.93%\\n----------------\\nBTC-USD | Total P/L % :: 5.92%\\nWinrate 52.60% | 1016 Trades, 485 Winners, 437 Losers, 94 Breakeven\\nAll Trades | Avg Trade 0.01% | SD 0.52%\\nWinners | Avg Trade 0.30% | SD 0.40%\\nLosers | Avg Trade -0.32% | SD 0.50%\\nProfit Factor: 1.0425461208948326 | Gross Profit: 144.96% | Gross Loss: -139.04%\\n----------------\\nETH-USD | Total P/L % :: 7.28%\\nWinrate 49.40% | 1166 Trades, 536 Winners, 549 Losers, 81 Breakeven\\nAll Trades | Avg Trade 0.01% | SD 0.41%\\nWinners | Avg Trade 0.28% | SD 0.33%\\nLosers | Avg Trade -0.26% | SD 0.33%\\nProfit Factor: 1.0501686433778488 | Gross Profit: 152.35% | Gross Loss: -145.07%\\n----------------\\nBNB-USD | Total P/L % :: 30.02%\\nWinrate 53.90% | 1245 Trades, 629 Winners, 538 Losers, 78 Breakeven\\nAll Trades | Avg Trade 0.02% | SD 0.44%\\nWinners | Avg Trade 0.29% | SD 0.37%\\nLosers | Avg Trade -0.29% | SD 0.33%\\nProfit Factor: 1.1945617990587647 | Gross Profit: 184.34% | Gross Loss: -154.32%\\n----------------\\nLUNA1-USD | Total P/L % :: 50.68%\\nWinrate 54.18% | 2358 Trades, 1180 Winners, 998 Losers, 180 Breakeven\\nAll Trades | Avg Trade 0.02% | SD 0.49%\\nWinners | Avg Trade 0.35% | SD 0.34%\\nLosers | Avg Trade -0.36% | SD 0.39%\\nProfit Factor: 1.1393385435967547 | Gross Profit: 414.38% | Gross Loss: -363.70%\\n----------------\\nALGO-USD | Total P/L % :: -9.09%\\nWinrate 51.15% | 1242 Trades, 577 Winners, 551 Losers, 114 Breakeven\\nAll Trades | Avg Trade -0.01% | SD 0.55%\\nWinners | Avg Trade 0.34% | SD 0.36%\\nLosers | Avg Trade -0.38% | SD 0.52%\\nProfit Factor: 0.9563075557946746 | Gross Profit: 198.91% | Gross Loss: -208.00%\\n----------------\\nHEX-USD | Total P/L % :: 33.37%\\nWinrate 52.29% | 1392 Trades, 684 Winners, 624 Losers, 84 Breakeven\\nAll Trades | Avg Trade 0.02% | SD 0.44%\\nWinners | Avg Trade 0.31% | SD 0.34%\\nLosers | Avg Trade -0.29% | SD 0.32%\\nProfit Factor: 1.184087073294854 | Gross Profit: 214.64% | Gross Loss: -181.27%\\n----------------\\nAVAX-USD | Total P/L % :: 6.22%\\nWinrate 52.57% | 1235 Trades, 593 Winners, 535 Losers, 107 Breakeven\\nAll Trades | Avg Trade 0.01% | SD 0.49%\\nWinners | Avg Trade 0.34% | SD 0.31%\\nLosers | Avg Trade -0.37% | SD 0.43%\\nProfit Factor: 1.0316654781994263 | Gross Profit: 202.75% | Gross Loss: -196.53%\\n----------------\\nDOGE-USD | Total P/L % :: -1.66%\\nWinrate 49.95% | 1125 Trades, 514 Winners, 515 Losers, 96 Breakeven\\nAll Trades | Avg Trade -0.00% | SD 0.63%\\nWinners | Avg Trade 0.41% | SD 0.51%\\nLosers | Avg Trade -0.42% | SD 0.51%\\nProfit Factor: 0.9922566583758843 | Gross Profit: 212.92% | Gross Loss: -214.58%\\n----------------\\n\\n'"
      ]
     },
     "execution_count": 184,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\"\"\"\n",
    "QQQ | Total P/L % :: -4.67%\n",
    "Winrate 50.00% | 257 Trades, 71 Winners, 71 Losers, 115 Breakeven\n",
    "All Trades | Avg Trade -0.02% | SD 0.37%\n",
    "Winners | Avg Trade 0.36% | SD 0.24%\n",
    "Losers | Avg Trade -0.42% | SD 0.35%\n",
    "Profit Factor: 0.8439297984360714 | Gross Profit: 25.26% | Gross Loss: -29.93%\n",
    "----------------\n",
    "BTC-USD | Total P/L % :: 5.92%\n",
    "Winrate 52.60% | 1016 Trades, 485 Winners, 437 Losers, 94 Breakeven\n",
    "All Trades | Avg Trade 0.01% | SD 0.52%\n",
    "Winners | Avg Trade 0.30% | SD 0.40%\n",
    "Losers | Avg Trade -0.32% | SD 0.50%\n",
    "Profit Factor: 1.0425461208948326 | Gross Profit: 144.96% | Gross Loss: -139.04%\n",
    "----------------\n",
    "ETH-USD | Total P/L % :: 7.28%\n",
    "Winrate 49.40% | 1166 Trades, 536 Winners, 549 Losers, 81 Breakeven\n",
    "All Trades | Avg Trade 0.01% | SD 0.41%\n",
    "Winners | Avg Trade 0.28% | SD 0.33%\n",
    "Losers | Avg Trade -0.26% | SD 0.33%\n",
    "Profit Factor: 1.0501686433778488 | Gross Profit: 152.35% | Gross Loss: -145.07%\n",
    "----------------\n",
    "BNB-USD | Total P/L % :: 30.02%\n",
    "Winrate 53.90% | 1245 Trades, 629 Winners, 538 Losers, 78 Breakeven\n",
    "All Trades | Avg Trade 0.02% | SD 0.44%\n",
    "Winners | Avg Trade 0.29% | SD 0.37%\n",
    "Losers | Avg Trade -0.29% | SD 0.33%\n",
    "Profit Factor: 1.1945617990587647 | Gross Profit: 184.34% | Gross Loss: -154.32%\n",
    "----------------\n",
    "LUNA1-USD | Total P/L % :: 50.68%\n",
    "Winrate 54.18% | 2358 Trades, 1180 Winners, 998 Losers, 180 Breakeven\n",
    "All Trades | Avg Trade 0.02% | SD 0.49%\n",
    "Winners | Avg Trade 0.35% | SD 0.34%\n",
    "Losers | Avg Trade -0.36% | SD 0.39%\n",
    "Profit Factor: 1.1393385435967547 | Gross Profit: 414.38% | Gross Loss: -363.70%\n",
    "----------------\n",
    "ALGO-USD | Total P/L % :: -9.09%\n",
    "Winrate 51.15% | 1242 Trades, 577 Winners, 551 Losers, 114 Breakeven\n",
    "All Trades | Avg Trade -0.01% | SD 0.55%\n",
    "Winners | Avg Trade 0.34% | SD 0.36%\n",
    "Losers | Avg Trade -0.38% | SD 0.52%\n",
    "Profit Factor: 0.9563075557946746 | Gross Profit: 198.91% | Gross Loss: -208.00%\n",
    "----------------\n",
    "HEX-USD | Total P/L % :: 33.37%\n",
    "Winrate 52.29% | 1392 Trades, 684 Winners, 624 Losers, 84 Breakeven\n",
    "All Trades | Avg Trade 0.02% | SD 0.44%\n",
    "Winners | Avg Trade 0.31% | SD 0.34%\n",
    "Losers | Avg Trade -0.29% | SD 0.32%\n",
    "Profit Factor: 1.184087073294854 | Gross Profit: 214.64% | Gross Loss: -181.27%\n",
    "----------------\n",
    "AVAX-USD | Total P/L % :: 6.22%\n",
    "Winrate 52.57% | 1235 Trades, 593 Winners, 535 Losers, 107 Breakeven\n",
    "All Trades | Avg Trade 0.01% | SD 0.49%\n",
    "Winners | Avg Trade 0.34% | SD 0.31%\n",
    "Losers | Avg Trade -0.37% | SD 0.43%\n",
    "Profit Factor: 1.0316654781994263 | Gross Profit: 202.75% | Gross Loss: -196.53%\n",
    "----------------\n",
    "DOGE-USD | Total P/L % :: -1.66%\n",
    "Winrate 49.95% | 1125 Trades, 514 Winners, 515 Losers, 96 Breakeven\n",
    "All Trades | Avg Trade -0.00% | SD 0.63%\n",
    "Winners | Avg Trade 0.41% | SD 0.51%\n",
    "Losers | Avg Trade -0.42% | SD 0.51%\n",
    "Profit Factor: 0.9922566583758843 | Gross Profit: 212.92% | Gross Loss: -214.58%\n",
    "----------------\n",
    "\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fixed Trades | Total P/L % :: 66.01%\n",
      "Winrate 52.43% | 3878 Trades, 1847 Winners, 1676 Losers, 355 Breakeven\n",
      "All Trades | Avg Trade 0.02% | SD 0.49%\n",
      "Winners | Avg Trade 0.34% | SD 0.38%\n",
      "Losers | Avg Trade -0.33% | SD 0.39%\n",
      "Profit Factor: 1.1190124798731227 | Gross Profit: 620.64% | Gross Loss: -554.63%\n"
     ]
    }
   ],
   "source": [
    "fixed_trades = []\n",
    "for day in history.keys():\n",
    "    _, pnls, pnls_dict, trades, trades_dict = history[day]\n",
    "    for key in trades_dict.keys():\n",
    "        if \"XRP-USD\" in key or \"ADA-USD\" in key:\n",
    "            continue\n",
    "        fixed_trades += trades_dict[key]\n",
    "stats(np.array(fixed_trades), \"Fixed Trades\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "HEX-USDBNB-USD | 3.02\n",
      "HEX-USDLUNA1-USD | 2.51\n",
      "ETH-USDLUNA1-USD | 2.36\n",
      "AVAX-USDBNB-USD | 2.01\n",
      "BTC-USDETH-USD | 1.90\n",
      "ADA-USDLUNA1-USD | 1.86\n",
      "BTC-USDBNB-USD | 1.70\n",
      "DOGE-USDETH-USD | 1.62\n",
      "BNB-USDALGO-USD | 1.61\n",
      "DOGE-USDADA-USD | 1.58\n",
      "ALGO-USDBNB-USD | 1.56\n",
      "XRP-USDETH-USD | 1.55\n",
      "DOGE-USDALGO-USD | 1.46\n",
      "ADA-USDHEX-USD | 1.39\n",
      "XRP-USDADA-USD | 1.30\n",
      "BNB-USDBTC-USD | 1.28\n",
      "BTC-USDADA-USD | 1.25\n",
      "LUNA1-USDETH-USD | 1.18\n",
      "BNB-USDETH-USD | 1.18\n",
      "ETH-USDALGO-USD | 1.17\n",
      "ALGO-USDHEX-USD | 1.12\n",
      "HEX-USDALGO-USD | 1.11\n",
      "XRP-USDLUNA1-USD | 1.10\n",
      "LUNA1-USDADA-USD | 1.05\n",
      "AVAX-USDALGO-USD | 1.01\n"
     ]
    }
   ],
   "source": [
    "inds = np.argsort(-np.array(pnls))\n",
    "symbs = [*pair_symbols]\n",
    "for i in range(25):\n",
    "    ind = inds[i]\n",
    "    print(f\"{symbs[ind]} | {pnls[ind]:.2f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "XRP-USDAVAX-USD | -1.19\n",
      "QQQDOGE-USD | -1.19\n",
      "QQQBTC-USD | -1.12\n",
      "BTC-USDQQQ | -1.05\n",
      "AVAX-USDXRP-USD | -1.00\n",
      "XRP-USDQQQ | -0.94\n",
      "AVAX-USDDOGE-USD | -0.90\n",
      "ALGO-USDLUNA1-USD | -0.83\n",
      "BNB-USDXRP-USD | -0.76\n",
      "ALGO-USDADA-USD | -0.68\n",
      "AVAX-USDHEX-USD | -0.67\n",
      "DOGE-USDHEX-USD | -0.67\n",
      "DOGE-USDQQQ | -0.66\n",
      "LUNA1-USDDOGE-USD | -0.55\n",
      "BTC-USDHEX-USD | -0.42\n",
      "ALGO-USDDOGE-USD | -0.41\n",
      "ALGO-USDXRP-USD | -0.36\n",
      "BTC-USDXRP-USD | -0.33\n",
      "QQQXRP-USD | -0.32\n",
      "HEX-USDBTC-USD | -0.29\n",
      "ADA-USDAVAX-USD | -0.27\n",
      "XRP-USDHEX-USD | -0.26\n",
      "BNB-USDQQQ | -0.26\n",
      "BTC-USDDOGE-USD | -0.24\n",
      "ALGO-USDAVAX-USD | -0.21\n"
     ]
    }
   ],
   "source": [
    "inds = np.argsort(np.array(pnls))\n",
    "symbs = [*pair_symbols]\n",
    "for i in range(25):\n",
    "    ind = inds[i]\n",
    "    print(f\"{symbs[ind]} | {pnls[ind]:.2f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "trades = np.array(trades)\n",
    "wins = trades[trades > 0]\n",
    "losses = trades[trades < 0]\n",
    "scratch = trades[trades == 0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "All Trades | Total P/L % :: 44.45%\n",
      "Winrate 56.40% | 437 Trades, 229 Winners, 177 Losers, 31 Breakeven\n",
      "All Trades | Avg Trade 0.10% | SD 0.52%\n",
      "Winners | Avg Trade 0.45% | SD 0.44%\n",
      "Losers | Avg Trade -0.33% | SD 0.27%\n",
      "Profit Factor: 1.7557963812847739 | Gross Profit: 103.26% | Gross Loss: -58.81%\n"
     ]
    }
   ],
   "source": [
    "def stats(trades, name):\n",
    "    def sub_stats(trades, name):\n",
    "        avg = np.mean(trades)\n",
    "        sd = np.std(trades)\n",
    "        print(f\"{name} | Avg Trade {avg:.2f}% | SD {sd:.2f}%\")\n",
    "    \n",
    "    wins = trades[trades > 0]\n",
    "    losses = trades[trades < 0]\n",
    "    scratch = trades[trades == 0]\n",
    "    \n",
    "    print(f\"{name} | Total P/L % :: {np.sum(trades):.2f}%\")\n",
    "    print(f\"Winrate {100*len(wins) / (len(trades) - len(scratch)):.2f}% | {len(trades)} Trades, {len(wins)} Winners, {len(losses)} Losers, {len(scratch)} Breakeven\")\n",
    "    sub_stats(trades, \"All Trades\")\n",
    "    sub_stats(wins, \"Winners\")\n",
    "    sub_stats(losses, \"Losers\")\n",
    "    print(f\"Profit Factor: {abs(np.sum(wins) / np.sum(losses))} | Gross Profit: {np.sum(wins):.2f}% | Gross Loss: {np.sum(losses):.2f}%\")\n",
    "\n",
    "stats(trades, \"All Trades\")\n",
    "    \n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "0b7367f57bf9904af5d15e1685ff3d84137deb646b78f1ef6564f95c07baf8ac"
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  },
  "varInspector": {
   "cols": {
    "lenName": 16,
    "lenType": 16,
    "lenVar": 40
   },
   "kernels_config": {
    "python": {
     "delete_cmd_postfix": "",
     "delete_cmd_prefix": "del ",
     "library": "var_list.py",
     "varRefreshCmd": "print(var_dic_list())"
    },
    "r": {
     "delete_cmd_postfix": ") ",
     "delete_cmd_prefix": "rm(",
     "library": "var_list.r",
     "varRefreshCmd": "cat(var_dic_list()) "
    }
   },
   "types_to_exclude": [
    "module",
    "function",
    "builtin_function_or_method",
    "instance",
    "_Feature"
   ],
   "window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}