Effectiveness of using quantified intermarket influence for predicting trading signals of stock markets

Authors:
Chandima D. Tilakaratne;Musa A. Mammadov;Sidney A. Morris
Affiliations:
University of Ballarat, Ballarat, Victoria, Australia and University of Colombo, Colombo, Sri Lanka;University of Ballarat, Ballarat, Victoria, Australia;University of Ballarat, Ballarat, Victoria, Australia
Venue:
AusDM '07 Proceedings of the sixth Australasian conference on Data mining and analytics - Volume 70
Year:
2007

Citing 4
Cited 1

Classifying trend movements in the MSCI U.S.A. capital market index—a comparison of regression, ARIMA and neural network methods

Computers and Operations Research - Special issue: artificial intelligence, evolutionary programming and operations research
Application of neural networks to an emerging financial market: forecasting and trading the Taiwan stock index

Computers and Operations Research - Special issue: Emerging economics
Time Dependent Directional Profit Model for Financial Time Series Forecasting

IJCNN '00 Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN'00)-Volume 5 - Volume 5
Quantification of Intermarket Influence Based on the Global Optimization and Its Application for Stock Market Prediction

AIDM '06 Proceedings of the International Workshop on on Integrating AI and Data Mining

Predicting Trading Signals of Stock Market Indices Using Neural Networks

AI '08 Proceedings of the 21st Australasian Joint Conference on Artificial Intelligence: Advances in Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper investigates the use of influence from foreign stock markets (intermarket influence) to predict the trading signals, buy, hold and sell, of the of a given stock market. Australian All Ordinary Index was selected as the stock market whose trading signals to be predicted. Influence is taken into account as a set of input variables for prediction. Two types of input variables were considered: quantified (weighted) input variables and their un-quantified counterparts. Two criteria was applied to determine the trading signals: one is based on the relative returns while the other uses the conditional probability that a given relative return is greater than or equals zero. The prediction of trading signals was done by Feedforward neural networks, Probabilistic neural networks and so called probabilistic approach which was proposed in past studies. Results suggested that using quantified intermarket influence as input variables to predict trading signals, is more effective than using their un-quantified counterparts.