Feed-Forward Neural Networks Regressions with Genetic Algorithms: Applications in Econometrics and Finance

Abstract:     
In this paper we examine feed-forward neural networks using genetic algorithms in the training process instead of error backpropagation algorithm. Additionally real encoding is preferred to binary encoding as it is more appropriate to find the optimum weights. We use learning and momentum rates for the weight updating as in the case of the error backpropagation algorithm. Some empirical examples as well as the programming routine in MATLAB are provided.

Number of Pages in PDF File: 25

Keywords: Feed-Forward Neural Networks, Genetic Algorithms, Time-Series, stock returns, inflation rate, gross domestic product, Forecasting, MATLAB

JEL Classification: C22, C45, C53, C63, G10

working papers series