The effect of kernel and bandwidth specification in geographically weighted regression models on the accuracy and uniformity of mass real estate appraisal (original) (raw)

Local governments have the responsibility of fairly and uniformly taxing the properties within their jurisdictions, and they must be held accountable for the taxes levied upon property owners. Therefore, it is imperative that residential property assessments be accurate, fair, and defensible. In recent years, great strides have been made in the advancement of mass-appraisal techniques such as automated valuation models (AVMs). It has long been understood that heterogeneity across geographic stratums hinders conventional ordinary-least-squares-based multiple-regression analysis (MRA) models from accurately capturing variables' true effects (Ball 1973; Berry and Bednarz 1975; Anselin and Griffith 1988). While spatial consideration in the form of dummy variables and distance coefficients can help improve models, these techniques may fail to fully correct for spatial autocor-relation, and parameter averages may be skewed or cancel each other out as a result (Berry and Bednarz 1975; ...