Performance evaluation of univariate time-series techniques for forecasting monthly rainfall data

Abstract In this article, the performance evaluation of four univariate time-series forecasting techniques, namely Hyndman Khandakar-Seasonal Autoregressive Integrated Moving Average (HK-SARIMA), Non-Stationary Thomas-Fiering (NSTF), Yeo-Johnson Transformed (YJNSTF) and Seasonal Naïve (SN) method, is carried out. The techniques are applied to forecast rainfall time series stations located in Kerala. It enables an assessment significant difference characteristics at various locations that influence relative accuracies models. Along with this, effectiveness transformation (YJT) improving accuracy models assessed. Rainfall 18 Kerala, India, starting from 1981 ending 2013, used. A classification system based on root mean square error (RMSE), absolute (MAE) Nash–Sutcliffe model efficiency coefficient (NSE) proposed find best model. HK-SARIMA YJNSTF performed well Western lowlands Eastern highlands. Central midlands, out 12 stations, indices 8 favour can be concluded more reliable for monthly all geographic regions state