An evolutionary algorithm based hybrid parallel framework for asia foreign exchange rate prediction

Abstract

This paper is about an evolutionary algorithm-based hybrid parallel model to enhance the prediction of Asia foreign exchange rate. This hybrid parallel model is made up of a trained adaptive linear combiner as linear model and a functional link artificial neural network (FLANN) as a non-linear model in parallel. To set the parameters of the non-linear model, differential evolution (DE) learning algorithm has been employed whereas the linear model has already been trained using LMS algorithm. We have primarily focused on Asia foreign exchange rate prediction for which, we have considered six Forex data set to validate the model and the study reveals that it outperforms than other models.