Re: st: Deleteing cases without warning in -hetgrot-

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 01:31 PM 3/8/2004 +0000, Clive Nicholas wrote:
> Unfortunately, there's a problem. After calling -hetgrot [panelvar]- (which
> runs the test), the ado-file deletes one or two observations from each
> panel in the dataset. As you can imagine, this is quite disconcerting, and
> it means that I have to -exit, clear- each time.
>
> I'm sure there's something in the code that causes this (and perhaps
> correctly, because of the degrees of freedom consumed in calculating the
> statistic). Can anybody see any way around this?
This part of the code would seem to be the obvious candidate:

> drop if e(sample)==0
If this is the culprit and you don't want the cases dropped, either (a) 
rewrite the code using if statements to always select those cases in 
e(sample), or (b) just reopen your data set at the end.

*
*   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/