Application of Neural Networks to an Emerging Financial Market: Forecasting and Trading the Taiwan Stock Index

Computers & Operations Research, Vol. 30, pp. 901-923

Posted: 28 Jan 2002

See all articles by Mark T. Leung

Mark T. Leung

University of Texas at San Antonio - Department of Management Science and Statistics

Hazem Daouk

Cornell University - School of Applied Economics and Management

An-Sing Chen

National Chung Cheng University - Department of Finance

Abstract

Although there exists some studies which deal with the issues of forecasting stock market index and development of trading strategies, most of the empirical findings are associated with the developed financial markets (e.g., U.S., U.K., and Japan). Currently, many international investment bankers and brokerage firms have major stakes in overseas markets. Given the economic success of Taiwan in the last two decades, the financial markets in this Asian country have attracted considerable global investments. Our study models and predicts the TSE Index using neural networks. Their performance is compared with that of parametric forecasting approaches, namely the Generalized Methods of Moments (GMM) and random walk. These rapidly growing financial markets are usually characterized by high volatility, relatively smaller capitalization, and less price efficiency, features which may hinder the effectiveness of those forecasting models developed for established markets. The good performance of the PNN suggests that the neural network models are useful in predicting the direction of index returns. Furthermore, PNN has demonstrated a stronger predictive power than both the GMM-Kalman filter and the random walk forecasting models. This superiority is partially attributed to PNN's ability to identify outliers and erroneous data. Compared to the other two parametric techniques examined in this study, PNN does not require any assumption of the underlying probability density functions of the class populations. The trading experiment shows that the PNN-guided trading strategies obtain higher profits than the other investment strategies utilizing the market direction generated by the parametric forecasting methods. In addition, the PNN-guided trading with multiple triggering thresholds is generally better than the one with single triggering thresholds. The multiple threshold version is able to consider the degree of certainty of a particular PNN classification and thereby reduce potential loss in the market.

Note: This is a description of the paper and not the actual abstract.

Keywords: emerging economy, index forecasting, trading strategy, neural networks, generalized gethods of moments (GMM)

Suggested Citation

Leung, Mark T. and Daouk, Hazem and Chen, An-Sing, Application of Neural Networks to an Emerging Financial Market: Forecasting and Trading the Taiwan Stock Index. Computers & Operations Research, Vol. 30, pp. 901-923. Available at SSRN: https://ssrn.com/abstract=277774

Mark T. Leung

University of Texas at San Antonio - Department of Management Science and Statistics ( email )

San Antonio, TX
United States

Hazem Daouk (Contact Author)

Cornell University - School of Applied Economics and Management ( email )

446 Warren Hall
Ithaca, NY 14853
United States
331-45-78-63-88 (Fax)

HOME PAGE: http://courses.cit.cornell.edu/hd35/

An-Sing Chen

National Chung Cheng University - Department of Finance ( email )

Chia-Yi, Taiwan 621
China
+011 886 5 272 0411 (Phone)
+011 886 5 272 0818 (Fax)

Register to save articles to
your library

Register

Paper statistics

Abstract Views
1,590
PlumX Metrics