Rationality Analysis for ARIMA Forecasts

This study is made to evaluate forecast efficiency by applying Rationality criterion for food price inflation, consumer price index general, GDP per capita and Money supply forecasts of Pakistan. It is therefore designed to analyze forecasting efficiency by using thirty three years annual data covering the period 1975 to 2008. We obtained forecasts from ARIMA(Auto Regressive Integrated Moving Average) model specification and select the most accurate forecast on the basis of well known forecasting accuracy techniques that are Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Theil’s Inequality Coefficient (TIC). Later on these forecasts are evaluated on the basis of rationality criterion. We found food price forecast are consistent, efficient and fulfilling the criteria given of weak and strong rationality, therefore they are reliable and correct to be used in policymaking and management decision where as other forecasts obtained are not passing all the criteria given except money supply.   Keywords: Food price Forecasts, Weak rationality, strong rationality, ARIMA forecasts