Trading rules for high-frequency financial data based on hybrid clustering

Author

Han-Min Ye ; Min Wang ; Ting-Liang Wang

Author_Institution

Sch. of Inf. Sci. & Technol., Guilin Univ. of Technol., Guilin, China

Volume

2

fYear

2011

fDate

26-28 July 2011

Firstpage

1191

Lastpage

1195

Abstract

A novel stock trading rule is proposed in this paper. The trading rule combines self-organizing map network and K-means clustering algorithm to discover the potential features of high frequency financial data at first, and then new trading strategies are proposed to assist investors´ transaction decision. The hybrid clustering based trading rule is applied to Hushen 300 Index of Chain, and compared its performances with other existing classical trading technique. Empirical research results give evidence that the newly proposed trading rule can be used to discover intraday patterns, and provide decision support to help stock investors gain more returns.

Keywords

financial data processing; pattern clustering; self-organising feature maps; stock markets; China; Hushen 300 Index; K-means clustering algorithm; high-frequency financial data; hybrid clustering; investor transaction decision; self-organizing map network; stock investment; stock trading rule; Clustering algorithms; Educational institutions; Indexes; Neurons; Stock markets; Strontium; Training; High frequency financial data; K-means; Self-organizing maps; Trading rules;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems and Knowledge Discovery (FSKD), 2011 Eighth International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-61284-180-9

Type

conf

DOI

10.1109/FSKD.2011.6019696

Filename

6019696