Using support vector machine with a hybrid feature selection method to forecast the direction of stock index
| Autor: | Zhi-Chao Yang, 楊智超 |
|---|---|
| Rok vydání: | 2008 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 96 The ability to forecast the trend of the stock markets is critical to analysts. Among the large array of approaches available for forecasting trend, support vector machine (SVM) are gaining in popularity. SVM classification is currently an active research area and successfully solves classification problems in many domains. SVM captures geometric characteristics of feature space without deriving weights of networks from the training data; it is capable of extracting the optimal solution with the small training set size. In this paper, we developed a prediction model based on support vector machine with a hybrid feature selection method to forecast stock index. This hybrid feature selection method combines filter and wrapper to select the optimal feature set from original feature set. The grid-search technique with 5-fold cross-validation is used to find out the best parameter value of kernel function of SVM. The experiment results show that the proposed SVM-based model provides a robust model with high prediction accuracy for forecasting NASDAQ Index. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|