Procedia Computer Science 31 (2014) 625 – 631
Available online at www.sciencedirect.com
1877-0509 © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of the Organizing Committee of ITQM 2014.
doi:10.1016/j.procs.2014.05.309
ScienceDirect
2nd International Conference on Information Technology and Quantitative Management,
ITQM 2014
Stock Price Prediction Based on SSA and SVM
WEN Fenghua
a*
, XIAO Jihong
b
, HE Zhifang
a
, GONG Xu
a
a. Business School of Central South University, Changsha, 410081, China
b. School of Economics and Management, Changsha University of Science
and Technology, Changsha, 410004 ,China
Abstract
This paper, using the singular spectrum analysis (SSA), decomposes the stock price into terms of the trend, the market
fluctuation, and the noise with different economic features over different time horizons, and then introduce these features
into the support vector machine (SVM) to make price predictions. The empirical evidence shows that, compared with the
SVM without these price features, the combination predictive methods——the EEMD-SVM and the SSA-SVM, which
combine the price features into the SVMs perform better, with the best prediction to the SSA-SVM.
© 2014 The Authors. Published by Elsevier B.V.
Selection and peer-review under responsibility of the Organizing Committee of ITQM 2014.
Keywords: Stock Price Series; Singular Spectrum Analysis; Support Vector Machine (SVM); Combination Predictive Methods
1. Introduction
Recently, the support vector machine
2
has been widely used in stock price predictions. However, as the
stock market is affected by economic, political, financial, social factors and noises, stock prices may have
different features over different time horizons. But few studies have introduced the price features into the SVM
to make price predictions. Zhang, et al
4
used the ensemble empirical mode decomposition (EEMD) to analyze
fundamental features of petroleum price series over different time horizons and pointed out that the
decomposed terms can be introduced into the SVM to make predictions. But the EEMD has some limitations in
the analysis of stock price series. The EEMD can not effectively extract noise from the price prediction, but the
impact of noise is prevalent in the stock market. Therefore, the EEMD can not catpure this feature well.
The SSA, a method for analyzing non-linear, non-stationary time series, was first proposed by Broomhead
and King
7
. The SSA is to get a series of singular values which contains the information of the original series
through the singular spectral decomposition (SVD). By analyzing singular values of different information, we
*Corresponding author: wfh@amss.ac.cn.
© 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and peer-review under responsibility of the Organizing Committee of ITQM 2014.