Bayesian methods of forecasting inventory investment in South Africa

Abstract

This paper develops a Bayesian Vector Error Correction Model (BVECM) for forecasting inventory investment in South Africa. The model is estimated using quarterly data on actual sales, production, unfilled orders, price levels and interest rates, for the period of 1978 to 2000. The out-of-sample-forecast accuracy obtained from the BVECM, over the forecasting horizon of 2001:1 to 2003:4, is compared with those generated from the Classical variant of the VAR and the VECM, the Bayesian VAR, and the ECM of inventory investment developed by Smith et al. (2006) for the South African economy. The BVECM with the most tight prior outperforms all the other models, except for a relatively tight BVAR. This BVAR model also correctly predicts the direction of change of inventory investment over the period of 2004:1 to 2006:3.