| Issue |
E3S Web Conf.
Volume 214, 2020
2020 International Conference on Energy Big Data and Low-carbon Development Management (EBLDM 2020)
|
|
|---|---|---|
| Article Number | 02047 | |
| Number of page(s) | 6 | |
| Section | Machine Learning and Energy Industry Structure Forecast Analysis | |
| DOI | https://doi.org/10.1051/e3sconf/202021402047 | |
| Published online | 07 December 2020 | |
Overview of Machine Learning for Stock Selection Based on Multi-Factor Models
1 College of Mathematics, Sichuan University, Chengdu, China
2 College of Mathematics, Sichuan University, Chengdu, China
3 College of Mathematics, Sichuan University, Chengdu, China
4 College of Mathematics, Sichuan University, Chengdu, China
a lihaoxuan_2002@126.com
b zhxy_1998@163.com
c liziyan99@163.com
d chunyuanzheng1999@163.com
In recent years, many scholars have used different methods to predict and select stocks. Empirical studies have shown that in multi-factor models, machine learning algorithms perform better on stock selection than traditional statistical methods. This article selects six classic machine learning algorithms, and takes the CSI 500 component stocks as an example, using 19 factors to select stocks. In this article, we introduce four of these algorithms in detail and apply them to select stocks. Finally, we back-test six machine learning algorithms, list the data, analyze the performance of each algorithm, and put forward some ideas on the direction of machine learning algorithm improvement.
© The Authors, published by EDP Sciences, 2020
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.