The Journal of Real Estate Finance and Economics
Volume 17, Issue 1, Article 3 (Abstract)
Title:	Predicting House Prices Using Multiple Listings Data
Author:	Robin A. Dubin
Abstract:	It is often necessary to accurately predict the price of a house between sales. One method of predicting house values is to use data on the characteristics of the area's housing stock to estimate a hedonic regression, using Ordinary Least Squares (OLS) as the statistical technique. The coefficients of this regression are then used to produce the predicted house prices. However, this procedure ignores a potentially large source of information regarding house prices: the correlations existing between the prices of neighboring houses. The purpose of this paper is to show how these correlations can be incorporated when estimating regression coefficients and when predicting house prices. The practical difficulties inherent in using a technique called kriging to predict house prices are discussed. The paper concludes with an example of the procedure using multiple listings data from Baltimore.
Keywords:	spatial autocorrelation, kriging, prediction