Shanghai Component Stock Index Forecasting Model Based on Data Mining
As it is known to all, many factors may have influence on the movement of stock index. In stock index forecasting, how many quantitative indicators should be introduced in order to obtain the best forecasting result? And is it true that more indicators tr
- PDF / 319,129 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 61 Downloads / 180 Views
Abstract As it is known to all, many factors may have influence on the movement of stock index. In stock index forecasting, how many quantitative indicators should be introduced in order to obtain the best forecasting result? And is it true that more indicators translate into higher forecasting accuracy? These issues have long been puzzling to researchers of stock index forecasting. In this paper, we carried out data mining on some quantitative indicators with influence on the movement of stock index, then we had short-term forecasting of Shanghai Component Stock Index with BP+GA model. Results of our research are as follows: forecasting with combination of indicators has better result than forecasting with single indicators; combinations of indicators through selection and optimization have the best result; more indicators introduced into forecasting model do not translate into higher accuracy. The results of our research in this paper demonstrate the necessity and significance of data mining in stock index forecasting. Keywords BP neural network index forecast Data mining
Genetic algorithm Forecast method Stock
1 Introduction There are many quantitative factors having influence on the movement of stock index. To solve the above problem, we first carried out data mining on the quantitative indicators with influence on the movement of stock index. Procedures W. Shen (&) X. Wu (&) T. Zhang School of Business and Administration, North China Electric Power University, Beijing, China e-mail: [email protected] X. Wu e-mail: [email protected]
Z. Wen and T. Li (eds.), Knowledge Engineering and Management, Advances in Intelligent Systems and Computing 278, DOI: 10.1007/978-3-642-54930-4_30, Springer-Verlag Berlin Heidelberg 2014
299
300
W. Shen et al.
of data mining are as follows: first, select a few frequently used technical indicators with higher correlation to the movement of stock index and started singleindicator forecasting; next, optimize indicators according to their performance in single-indicator forecasting and obtained an optimized combination for forecasting. In the process of data mining, we tried to observe if forecasting accuracy changes with the number of technical indicators. The next step is to select forecasting model. Statistical models (GARCH,SV) require complete samples and more information about their distribution; however, in China’s stock market, numbers of samples of listed companies change fast and the related information is not complete; therefore, the forecasting result would not be good if statistical models are used. Intelligent forecasting models, by comparison with statistical models, are more data-tolerant. GRAY model and SVM have their advantages in small sample forecasting but not suitable for large samples. Many researchers, when doing stock index forecasting, chose different indexes and different forecasting models. For example, Tsai and Hsiao [1] and Chu et al. [2] chose some quantitative indicators correlate with stock index fluctuation and introduced t
Data Loading...
