Creating and sharing knowledge for telecommunications

Fitness Function Evaluation for MA Trading Strategies based on Genetic Algorithms

Pinto, J. ; Neves, R. ; Horta, N.

Fitness Function Evaluation for MA Trading Strategies based on Genetic Algorithms , Proc Genetic and Evolutionary Computation Conf. - GECCO, Dublin, Ireland, Vol. NA, pp. 819 - 820, July, 2011.

Digital Object Identifier:

Download Full text PDF ( 419 KBs)

Abstract
This paper presents a new approach to optimize an investment strategy based on moving averages (MA). The proposed approach optimizes the entry and exit points, for both long and short positions, using a genetic algorithm (GA) kernel. This approach outperforms B&H strategy and explores alternative functions to the classical absolute return fitness function. The approach is demonstrated for major market indexes, such as, S&P 500, FTSE100, DAX30, NIKKEI225.