Intelligent algorithms for trading the euro-dollar in the foreign exchange market

In order to improve the profitability of Technical Analysis (here represented by Moving Average Crossover, DMAC), suitable optimization methods are proposed. Artificial Intelligence techniques can increase the profit performance of technical systems. In this paper, two intelligent trading systems are proposed. The first one makes use of fuzzy logic techniques to enhance the power of genetic procedures. The second system attempts to improve the performances of fuzzy system through Neural Networks. The target is to obtain good profits, avoiding drawdown situations, in applications to the DMAC rule for trading the euro-dollar in the foreign exchange market. The results show that the fuzzy system gives good profits over trading periods close to training period length, but the neuro-fuzzy system achieves the best profits in the majority of cases. Both systems show an optimal robustness to drawdown and a remarkable profit performance. In principle, the algorithms, described here, could be programmed on microchips. We use an hourly time series (1999-2012) of the Euro-Dollar exchange rate.