CEPII, Recherche et Expertise sur l'economie mondiale
Spatial Differencing: Estimation and Inference


Federico Belotti
Edoardo Di Porto
Gianluca Santoni

 Highlights :
  • Spatial differencing is a spatial data transformation pioneered by Holmes (1998) increasingly used to deal with omitted variable bias generated by local or site-specific unobservables in a "boundary-discontinuity" design setting.
  • However, the correlation between all differenced observations that share a common sample unit makes inference problematic.
  • We show that the dyadic-robust variance matrix estimator proposed by Cameron and Miller (2014) is, in general, a better solution compared to the most commonly used estimators.
  • A Stata command implementing the methods described in this paper is available. It can be installed from within Stata by typing net install sreg, from(http://www.econometrics.it/stata).

 Abstract :
Spatial differencing is a spatial data transformation pioneered by Holmes (1998) increasingly used to estimate causal effects with non-experimental data. Recently, this transformation has been widely used to deal with omitted variable bias generated by local or site-specific unobservables in a "boundary-discontinuity" design setting. However, as well known in this literature, spatial differencing makes inference problematic. Indeed, given a specific distance threshold, a sample unit may be the neighbor of a number of units on the opposite side of a specific boundary inducing correlation between all differenced observations that share a common sample unit. By recognizing that the spatial differencing transformation produces a special form of dyadic data, we show that the dyadic-robust variance matrix estimator proposed by Cameron and Miller (2014) is, in general, a better solution compared to the most commonly used estimators.

 Keywords : Spatial Differencing | Boundary Discontinuity | Robust Inference | Dyadic Data

 JEL : C12, C21
CEPII Working Paper
N°2017-10, June 2017

Texte intégral

Référence
BibTeX (with abstract),
plain text (with abstract),
RIS (with abstract)

Contact: 
 Domaines d'expertise

Compétitivité & Croissance
Retour