Testing for Homogeneity of Covariance

[A Ash]

Hi Stata experts!

I am trying to run the tests to satisfy assumptions for MANOVA. I have one sample with multiple IVs and DVs. My question is, do I need to run mvtest covariance Y1, Y2... by (group) (each IV) or is there some other or better way to do this?

[Marcos Almeida]

Hello "A" (please re-register with name and family name just by clicking on the "contact us" botton),

Here is a set of examples:http://www.stata.com/manuals13/mvmanova.pdf

Hopefully that helps.

Best,

Marcos

[Carlo Lazzaro]

A: (please consider Marcos' kind reminder about registering with full real name. Thanks).
As far as you query is concerned you can perform -mvtest- on covariaces using the group variable only:

mvtest covariances y1 y2 y3 y4, by(group)

.

In general; -manova- is robust (but to what extent who knows for sure) against departures from equality of group covariance matrices.
If I recall what follows correctly, some years ago a similar question popped up in Stata list and one of the repiies was to test the robustnes of the base case manova with another manova using observation ranks instead of raw observations.

[Marcos Almeida]

Carlo gave an excellent suggestion.

You may also want to further check your - manova - , and here you get some postestimation tools, with examples:http://www.stata.com/manuals13/mvman...postestimation

Best,

Marcos