Does Linear Regression save correct Mahalanobis distances?

Troubleshooting

Problem

I'd like to obtain Mahalanobis distances from each case in my data set to to the centroid for a set of variables in order to identify multivariate outliers. As described in Technote 1480128, I ran the Linear Regression procedure (Analyze->Regression->Linear) with the desired contributors to this distance as predictors and an arbitrary dependent variable. Here is the syntax command for my analysis: REGRESSION /MISSING LISTWISE /STATISTICS COEFF OUTS R ANOVA /DEPENDENT y /METHOD=ENTER x1 x2 x3 x4 x5 /CASEWISE PLOT(ZRESID) OUTLIERS(3) /SAVE MAHAL . When I validated these distances, it seemed as the Regression procedure has saved squared distances. That is, the square root sign seems to have been omitted from the Wikipedia equation: http://en.wikipedia.org/wiki/Mahalanobis_distance Please confirm or disconfirm whether SPSS renders SQUARED Mahalanobis distances instead of Mahalanobis distances?

[{"Product":{"code":"SSLVMB","label":"IBM SPSS Statistics"},"Business Unit":{"code":"BU048","label":"IBM Software"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"LOB76","label":"Data Platform"}}]

Log InLog in to view more of this document

This document has the abstract of a technical article that is available to authorized users once you have logged on. Please use Log in button above to access the full document. After log in, if you do not have the right authorization for this document, there will be instructions on what to do next.

Historical Number

83607

Was this topic helpful?

Document Information

Modified date:
16 April 2020

UID

swg21483596

Tips