The SVR Parameters Optimization For Stock'S Closing Price Forecast

DOI

10.2991/jimec-17.2017.53How to use a DOI?

Keywords

Support Vector Machines, stock prediction, Grid search, Genetic Algorithm, Particle Swarm Optimization

Abstract

Stock's closing price determines the price of the stock. So getting good prediction effect of stock's closing price is helpful to master the direction of the stock in some degree. In order to get better forecast effect in the stock's closing price to make use of SVR training and regression, fully considering the influence of parameters on the result of the experiment, mainly adopts the grid search parameters optimization, genetic algorithm and particle swarm optimization algorithm, three kinds of optimization algorithm of SVR parameters optimization. And 2275 rows historical data from April 9, 2007 to 2007 on August 10 are used for the simulation experiment. The experimental results show that the use of different optimization algorithms has much effect on the experimental results, and the regression results used the optimization algorithm are more accurate..

Copyright

© 2017, 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  - Tian Ye
PY  - 2017/10
DA  - 2017/10
TI  - The SVR Parameters Optimization For Stock'S Closing Price Forecast
BT  - Proceedings of the 2017 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017)
PB  - Atlantis Press
SP  - 244
EP  - 247
SN  - 2352-538X
UR  - https://doi.org/10.2991/jimec-17.2017.53
DO  - 10.2991/jimec-17.2017.53
ID  - Ye2017/10
ER  -