W: SPATIAL WEIGHT MATRIX DIVERSITY AND ITS IMPACT ON SPATIAL ANALYSIS RESULTS

Keywords: spatial analysis, spatial weight matrix, neighbourhood specification, spillover effects

Abstract

Spatial econometrics presents irreplaceable tool for regional analysis. Omitting additional information about geographical location of observed units could neglect some important influences. The spatial weight matrix W determining neighbourhood relations and degree of influence between observed units belongs to the main components of spatial analysis. Various specification approaches of this non-stochastic matrix could be applied. There is a commonly held belief that spatial regression models are sensitive to spatial weight structure. Some analytics consider it as a myth and points out incorrect interpretation of the model coefficients or misspecified models. Does it really matter what kind of specification is used? This contribution brings an empirical example of several approaches to neighbourhood specification and compares obtained results. According to findings of this analysis, especially spillover effects are incomparable. That confirms unequal performance of spatial structures. The W matrix should be built carefully at the beginning of each spatial analysis task.   

Downloads

Download data is not yet available.

Author Biography

Simona Mackova, University of Economics, Prague

Corresponding Author

simona.mackova1306@gmail.com

References

Anselin L. (2003). An Introduction to Spatial Regression Analysis in R. Illinois: Urbana-Champaign.
Retrieved from http://labs.bio.unc.edu/buckley/documents/anselinintrospatregres.pdf

Bivard R. (2016). Creating Neighbours. CRAN – R Project,
Retrieved from http://cran.r-project.org/web/packages/spdep/vignettes/nb.pdf.

Elhorst J.P. (2014). Spatial econometrics: from cross-section data to spatial panel. Heidelberg: Springer.

Formánek T. & Hušek R. (2015). The Czech Republic and its neighbors: Analysis of spatial macroeconomic dynamics. In 33th International Conference Mathematical Methods in Economics 2015, 190-195.
Retrieved from https://mme2015.zcu.cz/downloads/MME_2015_proceedings.pdf

Formánek T. & Hušek, R. (2016). On the stability of spatial econometric models: Application to the Czech Republic and its neighbors. In 34th International Conference Mathematical Methods in Economics 2016, 213-218.
Retrieved from
http://mme2016.tul.cz/conferenceproceedings/mme2016_conference_proceedings.pdf

Harris R., Moffat, J. & Kravtsova V. (2011). In Search of W. Spatial Econometric Analysis, 6(3), 249-270. https://doi.org/10.1080/17421772.2011.586721

LeSage J. & Pace R. K. (2009). Introduction to Spatial Econometrics. U.S.A.: CRC Press.

LeSage J. & Pace R. K. (2014). The biggest myth in spatial econometrics. Econometrics, 2(4), 217-249. http://dx.doi.org/10.2139/ssrn.1725503
Published
2019-10-20