Geographically Weighted Regression (GWR) is a technique that extends the traditional regression framework by allowing spatial parameters to be explicitly estimated. This paper provides a brief description of the Geographically Weighted Regression used here to value the effect of residential housing refurbishment in the City of Kaohsiung (Taiwan). The GWR results are then compared to a standard hedonic pricing estimation model applied to the same data set. What is intended here is to illustrate the use of a better tool for the identification of the spatial price impact of housing improvement investments in the metropolitan area. More generally, the paper confirms that spatial-adaptable models are required to measure the impact of investments in mixed and fuzzy goods.