Test performance and reliability of machine learning models from Stacking and Deep Learning for Stock Prediction.
- Deep Feed-forward Auto-Encoder Neural Network to reduce dimension + Deep Recurrent Neural Network + ARIMA + Extreme Boosting Gradient Regressor
- Adaboost + Bagging + Extra Trees + Gradient Boosting + Random Forest + XGB
- LSTM Recurrent Neural Network
- Encoder-Decoder Feed-forward + LSTM Recurrent Neural Network
- LSTM Bidirectional Neural Network
- 2-Path LSTM Recurrent Neural Network
- GRU Recurrent Neural Network
- Encoder-Decoder Feed-forward + GRU Recurrent Neural Network
- GRU Bidirectional Neural Network
- 2-Path GRU Recurrent Neural Network
- Vanilla Recurrent Neural Network
- Encoder-Decoder Feed-forward + Vanilla Recurrent Neural Network
- Vanilla Bidirectional Neural Network
- 2-Path Vanilla Recurrent Neural Network
- LSTM Sequence-to-Sequence Recurrent Neural Network
- LSTM with Attention Recurrent Neural Network
- LSTM Sequence-to-Sequence with Attention Recurrent Neural Network
- LSTM Sequence-to-Sequence Bidirectional Recurrent Neural Network
- LSTM Sequence-to-Sequence with Attention Bidirectional Recurrent Neural Network
- LSTM with Attention Scaled-Dot Recurrent Neural Network
- LSTM with Dilated Recurrent Neural Network
- Only Attention Neural Network
- Multihead Attention Neural Network
- LSTM with Bahdanau Attention
- LSTM with Luong Attention
- LSTM with Bahdanau + Luong Attention
# buy_stock(pred, df.Close,initial_state=1,delay=4,initial_money=10000,max_buy=3,max_sell=100)
day 0: buy 3 units at price 2306.100036, total balance 7693.899964
day 11, sell 3 units at price 2313.689940, investment 0.329123 %, total balance 10007.589904,
day 21: buy 3 units at price 2251.500000, total balance 7756.089904
day 27, sell 3 units at price 2367.810060, investment 5.165892 %, total balance 10123.899964,
day 37: buy 3 units at price 2374.649964, total balance 7749.250000
day 45, sell 3 units at price 2419.950072, investment 1.907654 %, total balance 10169.200072,
day 54: buy 3 units at price 2457.929994, total balance 7711.270078
day 65, sell 3 units at price 2420.909913, investment -1.506149 %, total balance 10132.179991,
day 77: buy 3 units at price 2485.920045, total balance 7646.259946
day 86, sell 3 units at price 2516.039979, investment 1.211621 %, total balance 10162.299925,
day 97: buy 3 units at price 2443.289979, total balance 7719.009946
day 109, sell 3 units at price 2470.049928, investment 1.095242 %, total balance 10189.059874,
day 120: buy 3 units at price 2622.750000, total balance 7566.309874
day 126, sell 3 units at price 2781.390015, investment 6.048614 %, total balance 10347.699889,
day 144: buy 3 units at price 2900.850036, total balance 7446.849853
day 152, sell 3 units at price 2860.200072, investment -1.401312 %, total balance 10307.049925,
day 158: buy 3 units at price 2878.350036, total balance 7428.699889
day 168, sell 3 units at price 2720.070006, investment -5.498985 %, total balance 10148.769895,
day 180: buy 3 units at price 2941.020081, total balance 7207.749814
day 191, sell 3 units at price 2780.369934, investment -5.462395 %, total balance 9988.119748,
day 201: buy 3 units at price 2774.070006, total balance 7214.049742
day 210, sell 3 units at price 2785.350036, investment 0.406624 %, total balance 9999.399778,
day 218: buy 3 units at price 2760.869934, total balance 7238.529844
day 227, sell 3 units at price 2848.500000, investment 3.174002 %, total balance 10087.029844,
day 241: buy 3 units at price 2978.429994, total balance 7108.599850
day 251, sell 3 units at price 3076.500000, investment 3.292675 %, total balance 10185.099850,
total gained 185.099850, total investment 1.850998 %
LSTM Recurrent Neural Network
LSTM Bidirectional Neural Network
2-Path LSTM Recurrent Neural Network
Deep Feed-forward Auto-Encoder Neural Network to reduce dimension + Deep Recurrent Neural Network + ARIMA + Extreme Boosting Gradient Regressor
LSTM Sequence-to-Sequence Recurrent Neural Network
LSTM Sequence-to-Sequence with Attention Recurrent Neural Network
LSTM Sequence-to-Sequence with Attention Bidirectional Recurrent Neural Network
Encoder-Decoder Feed-forward + LSTM Recurrent Neural Network
Adaboost + Bagging + Extra Trees + Gradient Boosting + Random Forest + XGB