Ensemble method based on ANFIS-ARIMA for rainfall prediction

This paper proposed an ensemble method based on ANFIS (Adaptive Neuro Fuzzy Inference System) and ARIMA (Autoregressive Integrated Moving Average) for forecasting monthly rainfall data at certain area in Indonesia, namely Pujon and Wagir area. The averaging method was implemented to find an ensemble forecast from ANFIS and ARIMA models. In this study, Gaussian, Gbell, and Triangular function are used as membership function in ANFIS. The forecast accuracy is compared to the best individual ARIMA and ANFIS. Based on root of mean square errors (RMSE) at testing datasets, the results show that an individual ANFIS method yields more accurate forecast in monthly Pujon's rainfall data, whereas ARIMA model yields better forecast in monthly Wagir's rainfall data. In general, these results in line with M3 competition results that more complicated model not always yield better forecast than simpler one.