> -----Original Message-----
> From: Stephen P. Jenkins [mailto:[email protected]]
> Sent: Thursday, March 27, 2003 10:19 AM
> To: [email protected]
> Subject: st: setting e(sample)
>
>
> Advice please about how to set the e(sample) function
> within an e class program.
>
> -ereturn b V post, esample(varname)- in version 8, and
> -est b V post, esample(varname)- in version 7,
> are the commands that the manuals refer to for setting e(sample).
>
> However both appear to assume that the program has created a
> coefficient vector ("b") and varcov matrix ("V").
> My problem is that my program does not create these (the
> estimates are
> built up in a non-regression based way).
> Is there anyway of creating e(sample) then?
>
> Stephen
A worst-case approach would be to create a "fake" b and V. This is the
strategy used by -svymean- and companion commands, when the -complete-
option is not specified.
After a call to -svymean- with the available option, for example, b and
V are as follows:
. mat list e(b)
symmetric e(b)[1,1]
dummy
y1 0
. mat list e(V)
symmetric e(V)[1,1]
dummy
dummy 0
--Nick Winter
*
* 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/
