Using Neural Networks to Enhance Technical Trading Rule Returns: A Case with KLCI

Abstract

In this paper, we test the profitability of technical trading rules which are enhanced by the use of neural networks on the Kuala Lumpur Composite Index (KLCI), a proxy of the Malaysian stock market traded in Bursa Malaysia. The profitable returns on KLCI from 2/1/2008 to 31/12/2014 offer a piece of evidence on the ability of technical trading rules using neural networks to outperform the buy-and-hold threshold benchmark. The test results here suggest that it is worthwhile to investigate, design and develop more robust machine learning algorithms, like neural networks enhanced technical indicators that enhance portfolio returns. The conclusion that can be drawn from this research work is that neural network may be used as tools in technical analysis for future price prediction. The findings from this work will interest all market participants, research analysis and fund managers who want to enhance their portfolio returns globally.