Prediction of Monthly Mean Temperature by SARIMA Model

Abstract

The monthly mean maximum and minimum temperature time series for four different stations of India were collected, modeled and forecasted between 1966-2015. In our forecasting the methods of the Box-Jenkins and Riddle-Smith seasonal auto regressive integrated moving-average is used with R software. It is demonstrated that obtained models are able to capture the dynamics of the time series data and to produce sensible forecasts.

Keywords: Forecast, Hydroclimatic, SARIMA, Time series.