| Title |
Gilaus mokymosi metodų, skirtų prognozuoti rinkų kintamumo indeksą akcijų rinkose, tyrimai / |
| Translation of Title |
Investigation of deep learning methods for volatility index prediction in stock markets. |
| Authors |
Šimašius, Marius |
| Full Text |
|
| Pages |
86 |
| Keywords [eng] |
VIX ; prediction ; deep learning network ; LSTM |
| Abstract [eng] |
The aim of this project is to analyze and investigate the possibilities of classical and deep learning methods to predict the market volatility index in stock markets. In this project, the methods and algorithms for forecasting the stock markets and the volatility index are analyzed. Classical ARMA / ARMAX / NARMA / NARMAX and deep learning models are applied to predict the change in VIX index for the following day and week, using VIX and other indices as model inputs. Experimental studies have shown that only a few models achieve sufficient accuracy to predict the interval of change in the VIX index (rising, stable, or falling) and the direction of change to be used as investment tools for investors or traders. It was also found that the mean absolute errors of the models are too high, so none of the developed deep learning models can be used as a reliable investment tool to predict the absolute value of the change in the VIX index. |
| Dissertation Institution |
Kauno technologijos universitetas. |
| Type |
Master thesis |
| Language |
Lithuanian |
| Publication date |
2022 |
