beathespread

ABSTRACT:Technical analysis is sometimes used in financial markets to assist traders make buying and selling decisions.

 

The success of technical analysis depends on how one interprets the available signals. Integration of human expertise into available models is considered to be essential for this purpose. Fuzzy systems could be used for developing decision models in which the experience of a trader can be incorporated in a natural way. In this paper, we examine a trading model that combines fuzzy logic and technical analysis to find patterns and trends in financial indices. The rule base of the fuzzy system is kept relatively straightforward for enhancing the interpretability of the model. The fuzzy model is optimised by using a genetic algorithm and historical data. The empirical results show that the proposed model is capable of generating higher risk-discounted returns in the out-of-sample periods compared to a buy-and-hold trading strategy with the exclusion of transaction costs. The results also indicate that the proposed system can outperform an existing portfolio allocation system in a financial institution, but the performance was not consistent over all considered

time periods.