The Index Prediction of Tehran Stock Exchange by Combining the Principal Components Analysis, Support Vector Regression and Particle Swarm Optimization

The prediction of future fluctuations in stock index can provide information about the future trend in the capital market. In order to increase the accuracy of the prediction of stock exchange index, this study used a combination of statistical methods and artificial intelligence. In this study, a hybrid stock index prediction model by utilizing principal component analysis (PCA), support vector regression (SVR) and particle swarm optimization (PSO) is proposed. In the proposed model, first, the PCA is used to deal with the nonlinearity property of the stock index data. The proposed model utilizes PCA to extract features from the observed stock index data. The features which can be used to represent underlying/hidden information of the data are then served as the inputs of SVR to build the stock index prediction model. Finally, PSO is applied to optimize the parameters of the SVR prediction model since the parameters of SVR must be carefully selected in establishing an effective and efficient SVR model. The findings show that preprocessing the data can decrease the prediction error of the model significantly.