Evaluating House Price Forecasts
House prices, unlike stock prices, appear to be predictable with some degree of accuracy. We use an autoregressive process to model the time series behavior of a city-wide house price index, and then produce one-quarter ahead forecasts for individual properties. Better real estate decisions require forecasting models with desirable properties for prediction errors (PEs). We propose that managers use a battery of tests to compare PEs; in particular, non-parametric smoothing of the empirical distribution of PEs can add important information to statistics that focus on first and second moments. The decision-making framework is fitted with housing transactions from Dade County, Florida, from 1976 through the second quarter of 1997. PEs from two forecasting models, hedonic and repeat sales, show some departure from the desirable properties of any one-step-ahead forecast. Also, both show some informational inefficiency, but the hedonic is more efficient than the repeat. Nonparametric smoothing shows that the hedonic method dominates the repeat over an important range of PEs; thus, a case can be made that many risk-averse managers would prefer a forecast based on the hedonic method.
Volume (Year): 24 (2002)
Issue (Month): 1 ()
|Contact details of provider:|| Postal: |
Web page: http://www.aresnet.org/
|Order Information:|| Postal: Diane Quarles American Real Estate Society Manager of Member Services Clemson University Box 341323 Clemson, SC 29634-1323|
Web: http://pages.jh.edu/jrer/about/get.htm Email:
When requesting a correction, please mention this item's handle: RePEc:jre:issued:v:24:n:1:2002:p:1-26. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (JRER Graduate Assistant/Webmaster)
If references are entirely missing, you can add them using this form.