The pegasus platform is used for price correlation on crypto-currency coins and tokens.
Using Make, everything can be run inside of a virtualenv. Be sure you have python 3.6.1
installed. We utilize the .python-version
file to specify our python version via pyenv. Once Python is setup properly, install dependencies:
$> make install
Running tests:
$> make test
New models are defined in the src/models
directory. You can inherity from Base model and implement the Pipeline
class. This is the starting point in processing the data.
Once implemented, you can define a database connection and a model schema for processing data.
model = Model(period=[2018, 2017, 2016, 2015, 2014, 2013], entities=[
{'slug': 'bitcoin', 'symbol': 'btc'},
{'slug': 'ethereum', 'symbol': 'eth'},
{'slug': 'litecoin', 'symbol': 'ltc'},
{'slug': 'monero', 'symbol': 'xmr'},
])
Pipeline(config, sqlite, model).process()
The motivation of this project was to learn deeply about volatility across a broad range of different currencies and to understand their specific risk and rewards.
The Markowitz Efficient frontier demonstrated below in the markowitz.py
model.