A multi-strategy differential evolution algorithm for financial prediction with single multiplicative neuron

Conference proceedings article

Authors/Editors

Strategic Research Themes

No matching items found.

Publication Details

Author list: Worasucheep C., Chongstitvatana P.

Publisher: Springer

Publication year: 2009

Volume number: 5864 LNCS

Issue number: PART 2

Start page: 122

End page: 130

Number of pages: 9

ISBN: 364210682X; 9783642106828

ISSN: 0302-9743

URL: https://www.scopus.com/inward/record.uri?eid=2-s2.0-76249097562&doi=10.1007%2f978-3-642-10684-2_14&partnerID=40&md5=ff6c55814fdf1b6fa0a5329ac20d1466

Languages: English-Great Britain (EN-GB)

View on publisher site

Abstract

This paper proposes a differential evolution (DE) algorithm that combines the strengths of multiple strategies together. The selection of strategy and control parameters for each individual happens every learning period. Thus the user gains the benefits of different strategies without difficult fine tuning of control parameters. The performance of the proposed MDE algorithm is evaluated on well-known benchmark functions and is superior to some other efficient and widely used variants of DE. In addition, MDE is applied to optimize both weights and biases of a single multiplicative neuron for prediction of DJIA with 3228 samples. Experiments show its better performance than other methods in learning ability and generalization. ฉ 2009 Springer-Verlag Berlin Heidelberg.

Keywords

Financial Prediction