Testing local versions of correlation coefficients


- Stars (0)


Owner: stam

Version: 1.0

Last Updated: 15-02-2016 1:30


The aim of this paper is to define and test local versions of standard correlation coefficients in statistical analysis. This research is motivated by the increasing number of applications using local versions of explanatory spatial data analysis methods such as local regression. One example of the latter is the Geographically Weighted Regression that allows for the researcher to check for the existence of spatial non-stationarity in the relationships between a geographic phenomenon and its determinants. In this paper, a local version of Pearson correlation coefficient is defined and applied to internal migration data. The results suggest that globally independent variables are not necessarily independent locally, thus the independence criterion may be violated when local regression analysis is performed. The work presented here is work-in-progress and will evolve along with the development of statistical software necessary for testing local versions of statistical tests.

Leave a comment

Your email address will not be published. Required fields are marked *