Application of SARIMA model for precipitation modelling driven by exogenous variables

The Seasonal Autoregressive Integrated Moving Average (SARIMA) model has gained popularity since its inception due to its ability to forecast seasonality. Usually, the SARIMA model captures the seasonality but does not consider the effect of the exogenous variable(s) in the seasonality process. Hence, this study aims to empirically introduce and implement the SARIMA-X model which can account for seasonality as well as the effects of influencing factors (X). Climate change has become the foremost global challenge facing human existence and the effect will be multifaceted with respect to social, economic and environmental challenges. This manuscript aims to forecast the precipitation time series data of Bangalore, India. The methodology employed in the analysis and modelling of precipitation series was the SARIMA-X model with exogenous variables temperature, relative humidity and surface pressure. In this manuscript, we have briefly discussed the SARIMA-X model along with its estimation procedure. The proposed model was diagnosed and the results showed that the model was adequate and parsimonious. The proposed model has compared with the traditional SARIMA model. The supremacy of using exogenous factors in seasonality modelling is concluded by this comparative study.