Forecasting Rainfall Pattern in Western Uttar Pradesh, India: A time series analysis using ARIMA

Forecasting Rainfall Pattern in Western Uttar Pradesh, India: A time series analysis using ARIMA

Abin Johnson, Apparasu Sruthi

Abstract

The prospect of adequate rainfall is a matter of concern for agricultural and environmental practitioners and policy makers. In this context, time series modeling and forecasting is being used to support these applications in forecasting rainfall patterns. The aim of this study is to make use of a well-used time series model ARIMA in forecasting annual rainfall in Uttar Pradesh (West) sub division, by making use of historical data of over 100 years (from 1901-2015). The data thus collected is transformed into a suitable model using R software, and it was found that ARIMA (0,1,3) is a model suitable for the given data set. As such, this model was used to forecast the pattern of annual rainfall in the region for twenty years (2016-2035) – which can be used in further research and policy making purposes. Also, the study theoretically notes the concepts of AR, MA, ARMA and SARIMA – where and how do we use them

Keywords: Rainfall in India, Uttar Pradesh, Time Series Data, ARIMA, forecasting