Author： Hodgson Douglas Slade Barrett Vorkink Keith
Publisher： Springer Publishing Company
ISSN： 0895-5638
Source： The Journal of Real Estate Finance and Economics, Vol.32, Iss.2, 2006-03, pp. : 151-168
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
Abstract
Constant-quality commercial indices generated by ordinary least squares may suffer an efficiency loss due to leptokurtosis caused by outliers in transactions data. When the subsequent nonnormality occurs, substantial improvement in index precision is obtained by estimating the hedonic model using a semiparametric adaptive estimator technique. When this method was applied to 1,846 office transactions that occurred in the Phoenix metropolitan area from January 1997 through June 2004, a substantial standard error reduction of approximately 9% was realized relative to ordinary least squares estimates. The difference in average returns between the semiparametric method and ordinary least squares was about 0.25% in each period, which represents a substantial increase in commercial property index precision.
Related content
Access to resources
Recommend
