ANALYSIS OF MONTHLY RAINFALL IN AMPARA DISTRICT USING SEASONAL ARIMA MODEL

Abstract

Rainfall is one of the foremost important phenomena of the natural system. The purpose of this research study is to fit a time series model to analyze and forecast future rainfall in Ampara district. Accordingly, this paper presents seasonal autoregressive integrated moving average model to analyze the monthly rainfall data in Ampara district. For the study purpose, monthly rainfall data from January 1991 to December 2020 were used. Using the Auto Correlation Function and Partial Auto Correlation Function plots the seasonality was determined. The seasonal autoregressive integrated moving average model was chosen based on the Akaike Information Criterion, Schwarz Information Criterion, Hannan-Quinn Criterion, Log Likelihood and Adjusted R-squared values. Based on the results obtained for the seasonal autoregressive integrated moving average models, SARIMA(0,0,2)(0,1,1)12 is selected as a best model for monthly rainfall for Ampara district.