Trading rules on stock markets using genetic network programming with subroutines

Author

Li, Jianhua ; Meng, QinBiao ; Yang, Yang ; Mabu, Shingo ; Wang, Yifei ; Hirasawa, Kotaro

Author_Institution

Grad. Sch. of Inf., Production & Syst., Waseda Univ., Kitakyushu, Japan

fYear

2010

fDate

18-21 Aug. 2010

Firstpage

3084

Lastpage

3088

Abstract

The purpose of this paper is to enhance the performance of Genetic Network Programming (GNP) to be used for creating trading rules on stocks, where a new method named GNP with Subroutines has been proposed. Compared to the conversional GNP, a new kind of node named subroutines node is added, which can call subprograms (subroutines) from GNP main programs. This reusable subroutines, which has judgment nodes and processing nodes working like small scale GNP, can evolve concurrently during the evolution of main GNP. In the simulations, the stock prices of different brands from 2001 to 2004 are used to test the effectiveness of the proposed method.

Keywords

commerce; genetic algorithms; network theory (graphs); stock markets; genetic network programming; reusable subroutine; stock market; stock price; trading rule; Algorithms; Economic indicators; Genetics; Indexes; Programming; Stock markets; Training; Automatically Defined Functions (ADFs); Genetic Network Programming; Subroutine; evolutionary algorithms; reinforcement learning; stock trading model; technical index;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE Annual Conference 2010, Proceedings of

Conference_Location

Taipei

Print_ISBN

978-1-4244-7642-8

Type

conf

Filename

5602828