Can Imposing Homogeity Restriction Improve the Forecast Accuracy of VECM? The Case of Meat Consumption in Urban Area of Iran

Using annual data of urban red meat, broiler, and fish consumptions over 1974-2008 period, this study compares the forecast performance of vector error correction models with and without imposing the homogeneity restriction in the cointegration space. To this end, out of sample forecasts for both restricted and unrestricted models were generated and evaluated based on the Mean Square Error (MSE) criterion. The test of equality of forecast errors indicated that there was a significant difference between the forecast errors of restricted and unrestricted models. This suggests that imposition of homogeneity restriction tends to improve the forecast accuracy when the restriction is not rejected.