Prediction of Temperature through SARIMA Model at Jabalpur in Madhya Pradesh

Abstract

The goal of this study was to develop a SARIMA (Seasonal Autoregressive Integrated Moving Average) model for weekly temperature prediction in the Jabalpur district of Madhya Pradesh. Weekly maximum and minimum temperature data from 1980 to 2018 were used to develop the model, while data from 2019 to 2021 were used to validate it in Jabalpur. SARIMA models were developed, and the best-fitting models with the highest accuracy were chosen for weekly maximum and minimum temperature prediction. The accuracy of the models was tested, and the results of various accuracy tests were reported in this paper.SARIMA (0,0,1)(2,1,0)52 and SARIMA(4,0,0)(1, 1, 0)52 were shown to produce better results. For SARIMA models, the MAE values are 1.60, 1.51, the RMSE is 2.113, 2.050, and the ACF1 value is 0.060, -0.001. The ability of the model to predict weekly temperature was determined to be good. The reported SARIMA models could be used to forecast the weekly temperature.