[R-sig-Geo] Mahalanobis scores for training data

[Lucas Fortini]

Dear Robert, 
I was replying to the thread because I have also encountered the same issue
that Gerardo mentions but believe that your recommendation does not address
the issue.
I have been using the mahal classifier function (Dismo package in r) in
several of my analyses and recently I have discovered that it seems to give
apparently wrong distance results for points that are identical to points
used in training of the classifier. I have recently posted a question on
stack overflow without realizing that this thread existed. In it I document
the odd behavior in multiple ways. 

http://stackoverflow.com/questions/24313694/mahalanobis-distance-based-classifier-leads-to-seemingly-wrong-scores-for-points

BTW, thanks very much for the work on the Dismo package. It is truly an
amazing tool set.

Best regards,
Lucas



--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Mahalanobis-scores-for-training-data-tp7585347p7586614.html
Sent from the R-sig-geo mailing list archive at Nabble.com.