Estimating the impact of air quality with a spatial hedonic: Geostatistical versus weight matrix approaches
Previous studies show that spatial methods are preferable to traditional ordinary least squares (OLS) hedonic method when location matters. This is particularly true for hedonic methods that include environmental characteristics such as air quality. Kim, Phipps, and Anselin (2003) use a spatial-weight matrix to model the presence of spatial dependence in the error term while Neill, Hassenzahl, and Assane (2007) use a geostatistical approach. Both papers report evidence that air quality matters but each spatial method is hampered by computational capabilities that restrict estimation to small data sets and it is ambiguous which of the two methods provide greater prediction relative to OLS. The purpose of this paper is to circumvent these limitations by using block bootstrapping method which is a form of Monte Carlo simulation that accounts for spatial dependence to explore the robustness of the predictive power of the two methods. Implications of the results for incorporating spatial effects into hedonic model are discussed.
Environmental Sciences | Statistical Methodology | Statistics and Probability
Use Find in Your Library, contact the author, or use interlibrary loan to garner a copy of the article. Publisher copyright policy allows author to archive post-print (author’s final manuscript). When post-print is available or publisher policy changes, the article will be deposited
Neill, H. R.,
Assane, D. D.,
Hassenzahl, D. M.
Estimating the impact of air quality with a spatial hedonic: Geostatistical versus weight matrix approaches.
North American Regional Science