Confounded Local Inference: Extending Local Moran Statistics to Handle Confounding

Levi John Wolf
Additional contact information
Levi John Wolf: University of Bristol

No kz5dq, SocArXiv from Center for Open Science

Abstract: Local geographic statistical analysis has long been of interest to scientists. In many cases, local statistics have been developed to identify spatial outliers --- areas that are markedly dissimilar from their surroundings --- or spatial clusters --- areas of strong similarity. Research into local spatial statistics experienced a step-change in the mid 1990s, which provided a large class of generalized estimators for local statistical analysis. The local Moran statistic is one commonly used local indicator of spatial association, obtained directly from the estimator for the global Moran statistic. However, this statistic (and local indicators more generally) have traditionally been univariate statistics. New developments provide a fully multivariate statistic for local spatial analysis: the multivariate Geary ratio. However, new arguments are needed to obtain a multivariable Moran statistic that can account for exogenous variation in its understanding of the local structure for spatial data. To do this, we return to the Moran Scatterplot as the critical analytical artifact for Moran-style analysis. Extending this concept, we develop a new method directly from a multivariable ``Moran-form'' spatial regression. We show the theoretical and empirical properties of this statistic, and contrast them to existing methods. Finally, we show how its use can change interpretations in an empirical analysis of rent in Bristol, England.

Date: 2022-10-07
New Economics Papers: this item is included in nep-geo and nep-ure
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://osf.io/download/633eff26d5b0100d7d1f9168/

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:osf:socarx:kz5dq

DOI: 10.31219/osf.io/kz5dq

Access Statistics for this paper

More papers in SocArXiv from Center for Open Science
Bibliographic data for series maintained by OSF (contact@cos.io).