Key points are not available for this paper at this time.
Significant amount of research have been carried out and various models have been developed by the researchers for weather forecasting. In this paper we present a comparative study of ARIMA (Auto-Regressive Integrated Moving Average) and ANFIS (Adaptive Network Based Fuzzy Inference System) models for forecasting the weather conditions in Dhaka, Bangladesh. Ten years weather data (from year 2000 to 2009), i.e., Maximum Temperature, Minimum Temperature, Humidity and Air Pressure are used in this research. We have compared the models with difference performance metric, for example, with Mean Absolute Error (MAE), Root Mean Square Error (RMSE), R-square error and the Sum of Square Error (SSE). Experimental results demonstrate that ARIMA has better performance compared to ANFIS. In this study, SPSS is used to carry out experiments on ARIMA model and Fuzzy Logic Toolbox in Matlab is used for ANFIS model.
Rahman et al. (Wed,) studied this question.