Applied Spatial Econometrics: Raising the Bar

Abstract

This paper places the key issues and implications of the new 'introductory' book on spatial econometrics by James LeSage & Kelley Pace (2009) in a broader perspective: the argument in favour of the spatial Durbin model, the use of indirect effects as a more valid basis for testing whether spatial spillovers are significant, the use of Bayesian posterior model probabilities to determine which spatial weights matrix best describes the data, and the book's contribution to the literature on spatiotemporal models. The main conclusion is that the state of the art of applied spatial econometrics has taken a step change with the publication of this book.Relever le niveau de l'économetrie spatial appliquée RÉ SUMÉ La pre´sente communication place les principales questions et implications du nouvel ouvrage d'introduction sur l'e´conome´tries spatiale de James LeSage & Kelley Pace (2009) dans un contexte plus ge´ne´ral: l'argument favorisant le mode`le spatial de Durbin, l'emploi d'effets indirects comme base plus valable pour e´valuer l'aspect significatif des de´versements spatiaux, l'emploi des probabilite´s d'un mode`le baysien poste´rieur pour e´valuer laquelle des matrices de poids spatiaux de´crit le mieux les donnes, et la contribution de l'ouvrage la documentation sur les mode`les spatio-temporels. La principale conclusion est qu'avec la publication de cet ouvrage, l'e´tat de l'art de l'e´conome´tries spatiale applique a effectue´un grand pas en avant.Alzar el nivel de la econometría espacial aplicada RÉ SUMÉ Este trabajo plantea las cuestiones e implicaciones clave del nuevo libro introductorio sobre econo´metra espacial de James LeSage & Kelley Pace (2009) dentro de una perspectiva ma´s amplia: el argumento a favor del modelo espacial Durbin, el uso de efectos indirectos como una base ma´s válida para poner a prueba si los desbordamientos espaciales son significativos, el uso de probabilidades posteriores bayesianas para descubrir que matriz de pesos espaciales describe mejor los datos, y la contribucio´n del libro a la biblio´grafa sobre modelos espaciotemporales. La principal conclusio´n es que la econometría espacial aplicada ma´s avanzada ha experimentado un cambio radical con la publicacio´n de este libro.

A spatial econometric analysis of land use efficiency in large and small municipalities

“…The spatial matrix is usually rowstandardized in a way that, when it pre-multiplies a variable, it returns the average value of the variable in the neighbours. Letting As suggested by Manski (1993) and confirmed by Elhorst (2010), the spatial model in equation (3) is the best strategy because it preserves the estimates from the omitted variable bias problem. However, 8 spatial heterogeneity turns out to be a crucial aspect of city size models.…”

Race to the debt trap? — Spatial econometric evidence on debt in German municipalities

“…In order to nd the most adequate spatial speci cation, we conduct various LM and LR tests along the lines of Elhorst (2012). The test results guide us to choose the SDM over the SLM.…”

“…In our main speci cation, we use a spatial Durbin model in a panel framework (Elhorst (2012)). Using this model, we present estimates for spatial interactions in debt and also 1 Note that the above arguments raise the question why politicians get away with everincreasing debt.…”

Housing market analysis using a hierarchical–spatial approach: the case of Belo Horizonte, Minas Gerais, Brazil

“…The empirical procedure for model selection was set similarly as the specific to general strategy proposed by Elhorst (2010). The methodological presentation begins with the Manski model, which is a general one:…”

“…Moreover, we wanted to choose between a spatial model with correlation in the errors or with endogenous and/or exogenous variables. Following a specific to general strategy of model selection (Elhorst, 2010), we initially estimated the SLM and applied the LM tests to its residues. The results suggested the use of the SAR models for all three groups of explanatory variables.…”