Title :
Short-term prediction of Shanghai composite index based on SVM
Author :
Wang, Xiaoyun ; Lin, Limin
Author_Institution :
Inst. of Manage. Sci. & Inf. Eng., Hangzhou Dianzi Univ., Hangzhou, China
Abstract :
Technical indicators are very important tools in the analysis of securities investment. Closing prices and volume of business are basic index, and they compose many complex technical index. In this paper, we represent the daily closing prices and daily volume of business as input vector, and construct 9 projects according different input vector. After 9 contrast experiments with support vector machines, we find that daily closing prices and daily volume of business have 3 days of time validity in predicting future stock price.
Keywords :
financial data processing; stock markets; support vector machines; Shanghai composite index; daily business volume; daily closing prices; future stock price; securities investment analysis; short-term prediction; support vector machines; Artificial neural networks; Business; Forecasting; Indexes; Kernel; Stock markets; Support vector machines; short-term prediction; support vector machine; technical indicator; time validity;
Conference_Titel :
Software Engineering and Service Sciences (ICSESS), 2010 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-6054-0
DOI :
10.1109/ICSESS.2010.5552390
