Training neural network with genetic algorithms for forecasting the stock price index

Author

Kai, Fu ; Wenhua, Xu

Author_Institution

Bank of China, Hebei, China

Volume

1

fYear

1997

fDate

28-31 Oct 1997

Firstpage

401

Abstract

The paper combines genetic algorithms (GA) with neural network (NN). It trains NN with GA and then predicts the stock price index with the trained network. By learning the special stock knowledge, it can find out the modes and relationship hidden in the abstract data. It can help shareholders and investment agencies to make wise decisions in the stock market so as to get more profits. The primary data are from Shanghai Stock Exchange from March 29, 1994 to August 1, 1994. The imitation result shows that the network is fit for short time prediction and it has high precision

Keywords

forecasting theory; genetic algorithms; investment; learning (artificial intelligence); neural nets; stock markets; Shanghai Stock Exchange; abstract data; genetic algorithms; investment agencies; neural network training; shareholders; short time prediction; special stock knowledge; stock price index forecasting; Content management; Economic forecasting; Expert systems; Genetic algorithms; Industrial training; Investments; Macroeconomics; Neural networks; Production; Stock markets;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-4253-4

Type

conf

DOI

10.1109/ICIPS.1997.672809

Filename

672809