“…There is clear evidence that many socio‐economic, geographic and demographic data used in the empirical analysis of regional science are characterised by the presence of spatial heterogeneity and spatial clustering. This can be found in some studies of the real estate market (e.g., Baumont, 2009; an de Meulen & Mitze, 2014; Cellmer et al, 2020; Wang et al, 2019), the location and/or economic performance of firms (e.g., Nilsson & Smirnov, 2017; Nilsson et al, 2019), the spatial distribution of poverty (e.g., Curtis et al, 2019; Oteng‐Abayie et al, 2022), local employment (e.g., Bradley et al, 2020), international migration (e.g., Hierro et al, 2013) and the regional distribution of per capita GDP, (e.g., Le Gallo & Ertur, 2003), among others. The inclusion of such information about the natural clustering of the data is captured through the inclusion in the weights of similarities or dissimilarities within neighbourhoods and between neighbourhoods in the predictor using either the local Moran's index or similarity measures between neighbourhoods via medians.…”