Establishment and Analysis of Multi-Factor Stock Selection Model Based on Support Vector Machine in CSI 300 Index Constituent Stocks Market

DOI

10.2991/aebmr.k.210319.139How to use a DOI?

Keywords

Quantitative Stock Selection, SVM, Multi-factor Model, Principal Component Analysis

Abstract

This paper uses the SVM (support vector machine) method to model the multi-factor stock selection and conducts research in Chinese Stock Market. The CSI 300 Index accounts for about 60% of the market value of Chinese Stock Market, we uses the principal component analysis for dimensionality reduction, reducing the number of original factors to 13, and the cumulative contribution rate reached 78.5372%, which reduced the complexity of SVM classification. In terms of model building, since the linear SVM method cannot be reasonably classified, this paper uses the radial basic kernel function and then classifies them, and obtains a stock selection model with strong effectiveness, which can beat the benchmark in all test sets. In terms of stock selection, we sort the stocks according to the sample values generated by the prediction of the model, and the reliability of the result obtained was high, which is the innovation of this paper.

Copyright

© 2021, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Changsheng Dou
AU  - Tengzhe Zhao
AU  - Ziheng Guo
PY  - 2021
DA  - 2021/03/22
TI  - Establishment and Analysis of Multi-Factor Stock Selection Model Based on Support Vector Machine in CSI 300 Index Constituent Stocks Market
BT  - Proceedings of the 6th International Conference on Financial Innovation and Economic Development (ICFIED 2021)
PB  - Atlantis Press
SP  - 746
EP  - 754
SN  - 2352-5428
UR  - https://doi.org/10.2991/aebmr.k.210319.139
DO  - 10.2991/aebmr.k.210319.139
ID  - Dou2021
ER  -