Skip to content

Latest commit

 

History

History
11 lines (10 loc) · 627 Bytes

README.md

File metadata and controls

11 lines (10 loc) · 627 Bytes
Raw

EMD-ALSTM-Multi-Factor-Stock-Profit-Prediction

Multi-Factor Stock Profit Prediction Using EMD-ALSTM

A deep learning solution to the 7th TipDM Cup National Data Mining Contest. Data acquisition method is through auto-trade api using desktop client software. The StrategyBase in strategy.py offers a rudimentary implementation of trading logic. Several of forcasting algorithms were implemented:

  1. Ordinary Linear Model (OLM): classic multi-factor model.
  2. Support Vector Machine (SVM)
  3. Multilayer Perception (MLP): generally a DNN was implemented.
  4. LSTM
  5. Attention-LSTM (ALSTM)
  6. Emperical Mode Decomposition ALSTM