Statistical Modeling and Forecasting of Food Grain in Effects on Public Distribution System: An application of ARIMA Model

Abstract

This study made an attempt to forecast and best fitted trend of Public Distribution System in India by using Box and Jenkins methodology of univariate Auto Regressive Integrated Moving Average (ARIMA) model. For empirical analysis a set of rice, wheat and total food grains for procurement, off take and stocks in PDS were considered for analysis with available annual data from 1972–2013 and forecasted values were estimated for 2017. The validity of the model were verified based on minimum value of AIC (Akaike Information Criterion), SBC (Schwarz's Bayesian Information Criterion), MAPE (Mean Absolute Percentage Error), maximum R2 etc. The work give a clear and complete view of Public Distribution System which takes a vital role in Food Security. ARIMA(2,1,1), ARIMA(1,1,2) and ARIMA(0,1,1) were found the best fitted models of total food grains for all three parameters. Predicted values of total food grains were observed 64.14, 58.24, 45.96 million tonnes in the period 2017 for all three parameters such as procurement, off-take, and stocks respectively.