Application of adaptive RPCL-CLP with trading system to foreign exchange investment

Author

Cheung, Yiu-Ming ; Lai, Helen Z H ; Xu, Lei

Author_Institution

Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong

Volume

4

fYear

1996

fDate

3-6 Jun 1996

Firstpage

2033

Abstract

In this paper, an adaptive rival penalized competitive learning and combined linear prediction (RPCL-CLP) model is applied to the forecast of stock price and exchange rate. As shown by the experimental results, this approach not only is better than Elman net and MA (q) models in the criterion of root mean square error, but also can bring in more returns in the trade between US dollar (USD) and German Deutschmark (DEM) with the association of a trading system. Moreover, whatever trading strategies with different risks are used in the trading system, adaptive RPCL-CLP can always keep the profits increasing as time goes through

Keywords

adaptive systems; financial data processing; forecasting theory; foreign exchange trading; investment; neural nets; unsupervised learning; Elman net; German Deutschmark; US dollar; adaptive rival penalized competitive learning; combined linear prediction; exchange rate forecasting; foreign exchange; investment; profits; root mean square error; stock price forecasting; trading system; Adaptive systems; Application software; Buffer storage; Clustering algorithms; Computer science; Exchange rates; Investments; Predictive models; Root mean square; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1996., IEEE International Conference on

Conference_Location

Washington, DC

Print_ISBN

0-7803-3210-5

Type

conf

DOI

10.1109/ICNN.1996.549214

Filename

549214