I have two data sets and want to compare/evaluate their degree of spatial association expressed in terms of the LISA statistic.

The local Moran's I statistic gives for each point a (pseudo) p-Value and a correlation coefficient. I want to make a statement like "The points of data set a are more locally correlated than the points of data set b". To do so I though of simply comparing the sum of the absolute values of the significant (p<0.05) correlation coefficients.

Do you think this is a valid approach? Alternatively it might be also reasonable to form the squared sum in order to give more weight to large coefficients.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.