Neuro-fuzzy hybridized model for seasonal rainfall forecasting: A case study in stock index forecasting

Abstract

The ensemble of statistics and mathematics has increased the accuracy of forecasting the Indian summer monsoon rainfall (ISMR) up to some extent. But due to the nonlinear nature of the ISMR, its forecasting accuracy is still below the satisfactory level. Mathematical and statistical models require complex computing power. Now a day, artificial neural networks (ANNs)-based models are used to forecast the ISMR. Various experiments signify that alone ANN cannot deal with the dynamic nature of the ISMR. So, in this chapter, we present a novel model based on the ensemble of ANN and fuzzy time series (FTS). This model is referred to as "Neuro-Fuzzy hybridized model for time series forecasting". The ISMR data set from the period 1901-1990 for the monsoon season (mean of June, July, August, and September) is considered for the experimental purpose. The forecasted results obtained for the training (1901-1960) and testing (1961-1990) data sets are then compared with existing models. The results clearly exhibit the superiority of our model over the considered existing models. The applicability of the proposed model has also been examined in the stock index data set.