Forecast Analysis of Food Price Inflation in Pakistan: Applying Rationality Criterion for VAR Forecast

Abstract

Forecast performance is considered to be a tart test of an econometric model. An accurate forecasting system is necessary for every industry to be able to take appropriate actions for future planning and planning creates a substantial need for forecasts. The purpose of this study is to evaluate forecast efficiency by using Rationality criterion of forecasts. It is therefore designed to analyze forecasting efficiency of food price inflation and consumer price index by using thirty three years quarterly data of Pakistan covering the period 1975 to 2008. Forecasts are obtained from VAR model specification. Four forecasting accuracy techniques, such as, Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE) and Theil‘s Inequality Coefficient (TIC) are used to be able to select the most accurate forecast from VAR. Later on these forecasts are evaluated on the basis of Rationality criterion defined. We found food price forecast are consistent, efficient and fulfilling the criteria of weak and strong rationality given. We propose that assessment of forecasts obtained by applying different criterion used will make them more reliable and correct to be used in policymaking and management decision.

Keywords: Food Price Forecasts, Weak Rationality, Strong rationality, Strict rationality