FORECASTING TOTAL FOOD GRAINS IN PAKISTAN

Abstract

Planning for inventory management of crops at government and enterprise level requires advance knowledge of likely crop output. Lack of such knowledge often results in sub-optimum decision about procurement and distribution of food grains, fixation of prices, imports and exports etc. This paper describes an empirical study of modelling and forecasting time series data of total food grains production in Pakistan. The Box-Jenkins ARIMA models and multiple regression model were used for forecasting. The diagnositing checking shows that ARIMA(2,2,2) is appropriate. Using ARIMA(2,2,2) and regression model eight yean ahead forecast was made. Forecasts by both the methods were compared and found reasonably close. These forecasts would be helpful for policy makers to plan the future requirements in terms of internal consumption and export to adopt appropriate measures in this regard.