Application of Adaptive Network-Based Fuzzy Inference System (ANFIS) with Genetic Algorithms in Crisis Periods Prediction

Abstract

In this paper discrete choice models, Logit and Probit are examined in order to predict the economic recession or expansion periods in USA. Additionally we propose an adaptive neurofuzzy inference system with triangular and Gaussian membership functions and genetic algorithms training optimization. We examine the in-sample period 1947-2005 and we test the models in the out-of sample period 2006-2009. The forecasting results indicate that the ANFIS model outperforms significant the Logit and Probit models in the out-of sample period. This indicates that neuro-fuzzy model provides a better and more reliable signal on whether or not a financial crisis will take place. We propose the use of both models, because with discrete choice models we can examine and investigate the effects of the inputs or the independent variables, while we can simultaneously use ANFIS for forecasting purposes. The wise option and the most appropriate scientific action is to combine both models and not taking only one of them and blindly supporting only one approach.