Skip to content

Latest commit

 

History

History
17 lines (15 loc) · 917 Bytes

File metadata and controls

17 lines (15 loc) · 917 Bytes

In this project I have looked at stacked ML models as shown below to predict stocked prices. This was part of a wider university project.

start.py will download all required data from yahoo finance. start.py contains all config data as well such as the target ticker, External comparison tickers and file paths for saved predictions.

          Historical Prices ───► ARIMA ───┐
                                          │
           Time Series Data ───►  LSTM ───┤
                                          │
           Time Series Data ───►  GRU  ───┤
                                          │
  Technical Indicators & Features ─►  RF ─┤
                                          │
                                          ▼
                                Meta Learner: XGBoost
                                  (Final Prediction)