Joel E. Pike <jpike@bus.wisc.edu> asks the following question, which has
also since turned up in a related form:
> I am running a linear regression using Stata 7/SE on a Windows 98 O/S,
> testing for heteroscedasticity. which is there, predicting cooksd, dropping
> outliers that exceed 4/_N, then running regress with the robust option.
> When I do this, there is no F or Prob F reported. I know that this is a
> Wald test with the Huber-White standard errors, but I can't find any
> documentation as to why this is missing. While I think I can access what
> Need to calculate this myself, I don't understand why it is missing.
Type
. help j_robustsingular
or from a Window-ed Stata simply click on the missing F-test (which should be
blue in this case) for an explanation. Basically, you probably have a
singularity in the estimated variance/covariance matrix which makes an overall
test a tricky thing. See the help file for an elaboration on this.
--Bobby
rgutierrez@stata.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/