Forecasting area and production of green gram in Odisha using ARIMA model

Abstract

A study was conducted on forecasting of area and production of green gram in Odisha. Box-Jenkins Autoregressive integrated moving average (ARIMA) time-series methodology was considered for forecasting of area and production of green gram. The different ARIMA models are judged on the basis of Autocorrelation Function (ACF) and Partial autocorrelation Function (PACF) at various lags The data from 1971-72 to 2006-07 are used for model building and from 2007-08 to 2015-16 used for successful cross-validation of the selected model on the basis of the absolute percentage error. The ARIMA models are fitted to the original time series data as well as the first difference data to check the stationarity. The possible ARIMA models are identified on the basis of significant coefficient of autoregressive and moving average components. The best fitted models are selected on the basis of low value of Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE). ARIMA (1,1,0) without constant found to be best fitted for area under green gram having absolute percentage error ranging from 0.34% to 10.63% in cross-validation of model. ARIMA (2,1,0) is the best fitted model for production of green gram having absolute percentage error ranges from 6.02% to 26.11% in cross-validation of the model. The best fitted ARIMA model has been used to forecast the area, production for the year 2016-17 to 2018-19. The model showed area forecast for the year 2018-19 to be about 895.75 thousand hectare with lower and upper limit 462.18 and 1329.31 thousand hectare respectively. The model also showed the forecasting in production of green gram for the year 2018-19 to be about 313.66 thousand tones with lower and upper limit 134.27 and 493.04 thousand tonnes respectively.