Spatial Autocorrelation in a Retail Context

Abstract

This paper describes and applies the weighted least squares (WLS) technique that corrects for spatial autocorrelation in the residuals of hedonic regressions. Most empirical studies to date have focused on spatial autocorrelation in the housing market, i.e., single family home valuation. This study focuses on mall stores within shopping centers, with an emphasis on retail site selection within the mall.