Forecasting of Pearl millet productivity in Gujarat under time series framework

Abstract

Pearl millet (Pennisetumglaucum) is one of the most extensively cultivated cereals in the world, ranking fourth after rice, wheat and sorghum. In Gujarat, for last 20 years, it has been noticed that the area under pearl millet production has been decreased gradually although productivity increase. The importance of pearl millet productivity forecasting is more relevant in semi-arid state like Gujarat where the precipitation is confirmed to short period of four months. In this paper, we have applied ARIMA model for forecasting of productivity of pearl millet of Gujarat. In the present study, time series data of pearl millet productivity (Kg./ha)of Gujarat for 52 years from 1960-61 to 2011-12 were collected from Directorate of Agriculture, Gandhinagar, Gujarat and partially from Directorate of Economics and statistics. The ARIMA model is validated on the basis of relative mean absolute prediction error (RMAPE), Mean absolute deviation (MAD) and root mean square error(RMSE) values. It may be noted that ARIMA (0, 1, 1) model performs quite satisfactorily as the RMAPE value is less than 6 percent