Forecasting of Islamic Stock Index in Turkey with Deep Learning Using Index-Based Features
The Islamic stock market is a dynamic platform that provides a suitable environment for investors to invest in shares that comply with Islamic law. Predicting the future price movements of the market provides significant advantages for investors such as reducing risk and increasing profits. As a con...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Istanbul University Press
2021-12-01
|
| Series: | Acta Infologica |
| Subjects: | |
| Online Access: | https://cdn.istanbul.edu.tr/file/JTA6CLJ8T5/F923B14399144D2F95102B5E3527727B |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The Islamic stock market is a dynamic platform that provides a suitable environment for investors to invest in shares that comply with Islamic law. Predicting the future price movements of the market provides significant advantages for investors such as reducing risk and increasing profits. As a consequence of developments in artificial intelligence applications, deep learning methods show superior success compared to other methods in predicting financial data. The Long Short Term Memory (LSTM) model, which can successfully model the complex relationship between input and output variables in a financial time series, attracts attention among deep learning methods. For this reason, the LSTM model was used in this study to forecast the Participation index, which represents the Islamic stock market in Turkey, with high accuracy. Macroeconomic factors or stock market technical indicators, which are widely used in the literature, were not used to determine the features that may directly affect the success of the model. Instead, following an index-based approach, the BIST 100 (XU100) index, CBOE volatility index (VIX), the gold volatility index (GVZ), and dollar index (DXY) were determined as the input variables of the forecast model. Thanks to this approach, many parameters are included in the model with a single index value and fewer input variables are used. Thus, the model is simplified and at the same time the predictive power of the model is increased. With the designed model, the Participation index was forecasted with 0.06, 0.08, 0.02, and 0.994 values in MAE, RMSE, MAPE, and R2 error functions, respectively. The main contribution of the study to the literature is that it is the first study in Turkey to use the LSTM model as a deep learning method in forecasting the Islamic stock index. The secondary contribution is the use of XU100, VIX, DXY, and GVZ parameters, which are index-based features, in the forecasting of the Islamic stock index. |
|---|---|
| ISSN: | 2602-3563 |