Could We Have Predicted the Recent Downturn in Home Sales in the Four U.S. Census Regions?

Abstract

This paper analyzes the ability of various models to forecast home sales in the four Census regions in the United States. We also use the models to predict the downturn in home sales in these regions over the 2004:Q4 to 2009:Q2 period. The findings reveal that, barring the South, there always exists a Bayesian model, which tends to outperform all other models in forecasting home sales over the out-of-sample horizon. In addition, when we expose our classical and 'optimal' Bayesian forecast models to predicting the peaks and declines in home sales, we find that barring the South again, the models did reasonably well in predicting the turning point exactly at 2005:Q3 or with a lead.