Research Article
Forecasting Foreign Exchange Volatility Using Deep Learning Autoencoder-LSTM Techniques
Table 6
Five error measures for each model in the EUVIX.
| | MSE | RMSE | MAE | MPE (%) | MAPE (%) |
| Period 1 | LSTM | 0.4104 | 0.6406 | 0.4893 | 2.5107 | 4.0954 | Autoencoder-LSTM | 0.2874 | 0.5361 | 0.3974 | −0.2569 | 3.3006 |
| Period 2 | LSTM | 4.5074 | 2.1231 | 1.4746 | 3.8257 | 19.3755 | Autoencoder-LSTM | 3.8284 | 1.9566 | 1.4937 | −6.4252 | 18.0702 |
| Period 3 | LSTM | 1.5205 | 1.2331 | 0.4530 | 0.5383 | 7.8089 | Autoencoder-LSTM | 1.3250 | 1.1511 | 0.5018 | −1.9384 | 8.7721 |
|
|