Karen Scheltema wrote in message <[EMAIL PROTECTED]>...
>What is the usual cutoff for saying the VIF is too high?

I don't see that there can be any general criterion for saying that
a VIF is too large.   A large value indicates collinearity between
predictor variables.   In some fields, this cannot be avoided.
I have one set of set for which most of the VIFs are in excess
of a million.   The data are from NIR spectroscopy, where this
is unavoidable.

If you do have large VIFs then make sure that your least-squares
software uses some form of orthogonal reduction.   If it uses
the normal equations, and hence squares the condition number,
then you could be in trouble.

>
>Karen Scheltema, M.A., M.S.
>Statistician
>HealthEast
>1700 University Ave W
>St. Paul, MN 55104
>(651) 232-5212   fax: (651) 641-0683
>

--
Alan Miller, Retired Scientist (Statistician)
CSIRO Mathematical & Information Sciences
Alan.Miller -at- vic.cmis.csiro.au
http://www.ozemail.com.au/~milleraj
http://users.bigpond.net.au/amiller/





===========================================================================
This list is open to everyone.  Occasionally, less thoughtful
people send inappropriate messages.  Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.

For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================

Reply via email to