Monday, February 13, 2017

On the specification of spatial models

One of best sentences I've read in an academic paper in years:
"Without divine intervention it is generally difficult to know with certainty which (if either) of the two above cases are true"
This is from Fotheringham et al (1998) on the specification of spatial models. I find this quite amusing but I must say this is one of the most well written and accessible articles on spatial econometric models I've come across so far. It's not by chance this paper has become a great reference on the topic with more than 500 citations.

I've only started reading more about spatial models recently. Here are four papers I would recommend to get started on the topic.
  • Anselin, L. (2002). Under the hood Issues in the specification and interpretation of spatial regression models. Agricultural Economics, 27(3), 247–267.
  • Fotheringham, A. S., Charlton, M. E., & Brunsdon, C. (1998). Geographically Weighted Regression: A Natural Evolution of the Expansion Method for Spatial Data Analysis. Environment and Planning A, 30(11), 1905–1927.
  • Florax, R. J. G. M., Folmer, H., & Rey, S. J. (2003). Specification searches in spatial econometrics: the relevance of Hendry’s methodology. Regional Science and Urban Economics, 33(5), 557–579. [thanks Leo Monasterio for the recommendation]
  • Páez, A., & Scott, D. M. (2005). Spatial statistics for urban analysis: A review of techniques with examples. GeoJournal, 61(1), 53–67.

This paper is particularly relevant to problem raised in the quote above:

  • Gibbons, S., & Overman, H. G. (2012). Mostly Pointless Spatial Econometrics?*. Journal of Regional Science, 52(2), 172–191




I just happened to like this plot.