A foreign exchange portfolio management mechanism based on fuzzy neural networks

The key in foreign exchange (Forex) trading is to pick the right currency to trade at the right time, primarily based on accurate forecast of future exchange rates. This paper presents a novel neuro-fuzzy approach in foreign exchange (Forex) portfolio management to pick the right pairs of currencies to buy and sell with optimized market timing. The proposed mechanism forecasts future BUY/SELL signals before matching these offsetting signals across different currencies to maximize trade returns. This mechanism makes use of fuzzy neural network (FNNs) as a forecasting tool, technical indicators such as moving averages and a novel Portfolio Trade Timing Optimization (PTTO) algorithm to produce an optimized BUY-SELL schedule for the Forex portfolio under management. Experimental results on real world Forex market data shows that the proposed mechanism yields significantly higher profits against various popular benchmarks.