The goal is to predict the price change of a security.
At each point in the above graph, we want to predict the prices after 1 min.
The following table shows the label for AAPL on 20170928.
Timestamp (EST) | Current Price | Price Difference |
---|---|---|
09:30:01.500000 | 153.905 | 0.125 |
09:30:01.550000 | 153.905 | 0.125 |
09:30:01.600000 | 153.9 | 0.13 |
09:30:01.650000 | 153.905 | 0.13 |
09:30:01.700000 | 153.905 | 0.13 |
InSample | OutSample | |
---|---|---|
MSE | 0.000242089 | 43997.8 |
MAE | 0.0118442 | 151.201 |
Original stdev | 0.0159 | 0.0202 |
Predicted stdev | 0.0108 | 179.1057 |
InSample | OutSample | |
---|---|---|
MSE | 0.000219 | 0.000557 |
MAE | 0.01101 | 0.01722 |
Original stdev | 0.0159 | 0.0202 |
Predicted stdev | 0.0050 | 0.0121 |
InSample | OutSample | |
---|---|---|
MSE | 0.000244 | 0.000412 |
MAE | 0.01113 | 0.01313 |
Original stdev | 0.0159 | 0.0202 |
Predicted stdev | 0.0020 | 0.0018 |
InSample | OutSample | |
---|---|---|
MSE | 0.000244 | 0.000412 |
MAE | 0.01114 | 0.01311 |
Original stdev | 0.0159 | 0.0202 |
Predicted stdev | 0.0020 | 0.0017 |
Algorithm | Stdev | OutSample MAE |
---|---|---|
Gradient Boosting | 0.0075 (0.0038) | 0.01583 |
AdaBoosting | 0.0037 (0.0058) | 0.01345 |
RandomForest | 0.0236 (0.0149) | 0.02638 |