Forecasting us home sales using bvar models and survey data on households' buying attitudes for homes

Abstract

This study uses Bayesian vector autoregressive models to examine the usefulness of survey data on households' buying attitudes for homes in predicting sales of homes. We find a negligible deterioration in the accuracy of forecasts of home sales when buying attitudes are dropped from a model that includes the price of homes, the mortgage rate, real personal disposable income, and the unemployment rate. This suggests that buying attitudes do not add much to the information contained in these variables. We also find that forecasts from the model that includes both buying attitudes and the aforementioned variables are similar to those generated from a model that excludes the survey data but contains the other variables. Additionally, the variance decompositions suggest that the gain from including the survey data in the model that already contains other economic variables is small.

Analysis of Consumers’ Perceptions of Buying Conditions for Houses

“…Several models have examined the effect of macroeconomic aggregates on the housing sector such as Kearl (1979), Follain (1981), Schwab (1983), Arnott (1987), Manchester (1987), Schwartz (1988), Smith et al (1988), Harris (1989), Megbolugbe et al (1991), Dua and Smyth (1995), Hanushek and Quigley (1982), Baffoe-Bonnie (1998), Chen and Patel (1998) and Ewing and Wang (2002).…”

“…From these responses, we construct an index of a good time to buy a house as follows: Variations of the index can be constructed as in Dua and Smyth (1995 …”

On the predictability of time-varying VAR and DSGE models

“…As shown in Harvey (1986), the time-domain treatment still goes through. The Kalman …lter may be applied to (26) and (27), the number of sets of observations needed to form an estimator of t , with …nite MSE matrix being the same as in the univariate case. The conditions for the …lter to converge to a steady state are an obvious generalization of the conditions in the univariate case.…”

Could we have predicted the recent downturn in the South African housing market?

“…Another approach is to use a near VAR, which specifies an unequal number of lags for the different equations. 6 Note, Dua and Smyth (1995), Dua and Miller (1996) and Dua et al (1999) used coincident and leading indexes in BVAR models to forecast home sales for the Connecticut and the overall US economy, respectively.…”