I wrote:
Using [Dmitry's code], I get missing values when one of
the x's predicts failure perfectly.
[snip listing]
Do you want this? It happens because the observations are dropped at the
execution of -probit-:
note: x2 != 1 predicts failure perfectly
x2 dropped and 63 obs not used
You can omit the -if e(sample)- if you want to avoid this. If you
need to keep the -if e(sample)- in the -adjust- command for [some other]
reason, then you can fix-up the predictor list up-front, prior to -probit-.
--------------------------------------------------------------------------------
That's muddled. There were two things going: a candidate predictor was
dropped and the perfectly predicted outcomes were omitted from the
regression on the remaining predictors. The latter is intended to avoid
biasing the remaining coefficients. My "fix-up" has the effect of
circumventing this, and could lead to bias in the remaining coefficients in
the context of a model that considers the dropped predictor.
Joseph Coveney
*
* 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/
