Evaluation of statistical models for forecasting area, production and productivity of groundnut in Gujarat

Groundnut is one of the most important oilseed crops in India. Forecasting is used to provide decision making and planning for the future effectively and efficiently. The study was carried out to estimate the trends of area, production and productivity of groundnut crop in Gujarat for the period of 1991-92 to 2019-20. The data were collected from the Directorate of Agriculture, Gandhinagar, Gujarat. The linear, quadratic and cubic models were fitted on original data along with moving average method. The ARIMA models were also developed to forecast area, production and productivity of groundnut. Among the fitted polynomial and ARIMA models, the suitable model was identified on basis of significance of regression coefficient, AR and MA, AIC and SBC values, adjusted R2, RMSE, MAE, normality and randomness test etc. The study revealed that, for the area, cubic model on five year moving average and for production and productivity, linear model on five year moving average were found best in Gujarat. The ARIMA(2,1,0) model wasfound suitable to forecast and explain the pattern of groundnut area in Gujarat. It was also observed that none of the ARIMA models tested was suitable for predicting groundnut production and productivity in Gujarat.