A hybrid SOFM-SVR with a filter-based feature selection for stock market forecasting

1. A hybrid SOFM-SVR with a filter-based feature selectionfor stock market forecasting Huang, C. L. & Tsai, C. Y. Expert Systems with Applications 2008

2. Introduction • Stock market price index prediction is regarded as a challenging task of the finance. • Support vector regression (SVR) has successfully solved prediction problems in many domains, including the stock market.

3. Introduction • filter-based feature selection to choose important input attributes • SOFM algorithm to cluster the training samples • SVR to predict the stock market price index • Using a real future dataset – Taiwan index futures (FITX) to predict the next day’s price index

4. Introduction • SOFM+SVR : to improve the prediction accuracy of the traditional SVR method and to reduce its long training time, • SOFM+SVR+filter-based feature selection : improvement in training time, prediction accuracy, and the ability to select a better feature subset is achieved.

5. SVR • Unlike pattern recognition problems where the desired outputs are discrete values (e.g., Boolean) • support vector regression (SVR) deals with ‘real valued’ functions

6. Self-organizing Feature Maps; SOFM

7. SOFM 1 2 3 4

8. Training the SOFM-SVR model • 1. Scaling the training set • 2.Clustering the training dataset • 3.Training the Individual SVR Models for Each Cluster

9. Training the SOFM-SVR model

10. Parameters Optimization • setting of the SVR parameters can improve the SVR prediction accuracy • Using RBF kernel and ε-insensitive loss function, three parameters, C, r, and ε, should be determined in the SVR model • The grid search approach is a common method to search for the C, r, and ε values.

11. Grid Search Approach

12. Evaluating the SOFM-SVR model with test set • Scale the test set based on the scaling equation according to the attribute rage of the training set • Find the cluster to which the test sample in the test set • Calculate the predicted value for each sample in the test set • Calculate the prediction accuracy for the test set

13. SOFM-SVR model

14. SOFM-SVR combined with filter-based feature selection • X is Certain input variable (i.e. feature) • Y is response variable (i.e. label) • n is the number of training samples

15. SOFM-SVR filter-based feature selection

16. Performance measures • Ai is the actual value of sample i • Fiis a predicted value of sample i • n is the number of samples.

17. Experimental data set

18. SOFM-SVR with various numbers of clusters in dataset #1

19. Accuracy measures with various numbers of clusters

20. Wilcoxon sign rank test Wilcoxon sign rank test on the prediction errors for the SOFM-SVR with various numbers of clusters

21. Results of SOFM-SVR using three clusters

22. Results of SOFM-SVR with selected features

23. Original Feature VS. Original Feature • Original Feature • Original Feature Wilcoxon sign rank test

24. Important Feature • MA10: 10-day moving average. • MACD9: 9-day moving average convergence/ divergence. • +DI10: directional indicator up. • -DI10: directional indicator down. • K10: 10-day stochastic index K • PSY10: 10-day psychological line. • D9: 9-day stochastic index D

25. Relative importance of the selected features

26. Wilcoxon sign rank test: SOFM-SVR vs. single SVR

27. MAPE comparison: SOFM-SVR vs. single SVRs.

28. Training time comparisons: SOFM-SVR vs. single SVRs.

29. Conclusion • Hybrid SOFM-SVR with filter based feature selection to improve the prediction accuracy and to reduce the training time for the financial daily stock index prediction • Further research directions are using optimization algorithms (e.g., genetic algorithms) to optimize the SVR parameters and performing feature selection using a wrapper-based approach that combines SVR with other optimization tools

30. Thank You