A Stock Market Prediction Method Based on Support Vector Machines (SVM) and Independent Component Analysis (ICA)
Hakob Grigoryan ()
Additional contact information
Hakob Grigoryan: University of Economic Studies, Bucharest, Romania
Database Systems Journal, 2016, vol. 7, issue 1, 12-21
The research presented in this work focuses on financial time series prediction problem. The integrated prediction model based on support vector machines (SVM) with independent component analysis (ICA) (called SVM-ICA) is proposed for stock market prediction. The presented approach first uses ICA technique to extract important features from the research data, and then applies SVM technique to perform time series prediction. The results obtained from the SVM-ICA technique are compared with the results of SVM-based model without using any pre-processing step. In order to show the effectiveness of the proposed methodology, two different research data are used as illustrative examples. In experiments, the root mean square error (RMSE) measure is used to evaluate the performance of proposed models. The comparative analysis leads to the conclusion that the proposed SVM-ICA model outperforms the simple SVM-based model in forecasting task of nonstationary time series.
Keywords: support vector machines; regression; independent component analysis; financial time series; stock prediction (search for similar items in EconPapers)
References: View references in EconPapers View complete reference list from CitEc
Citations View citations in EconPapers (1) Track citations by RSS feed
Downloads: (external link)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:aes:dbjour:v:7:y:2016:i:1:p:12-21
Access Statistics for this article
Database Systems Journal is currently edited by Ion Lungu
More articles in Database Systems Journal from Academy of Economic Studies - Bucharest, Romania Contact information at EDIRC.
Series data maintained by Adela Bara ().