Forecasting Real U.S. House Prices: Principal Components Versus Bayesian Regressions

Abstract

This paper analyzes the ability of principal component regressions and Bayesian regression methods under Gaussian and double-exponential prior in forecasting the real house prices of the United States (U.S.), based on a monthly dataset of 112 macroeconomic variables. Using an in-sample period of 1992:01 to 2000:12, Bayesian regressions are used to forecast real U.S. house prices at the twelve-months-ahead forecast horizon over the out-of-sample period of 2001:01 to 2004:10. In terms of the Mean Square Forecast Errors (MSFEs), our results indicate that a principal component regression with only one factor is best-suited for forecasting the real U.S. house prices. Among the Bayesian models, the regression based on the double exponential prior outperforms the model with Gaussian assumptions.