At 12:54 AM 5/25/2005 +0100, Clive Nicholas wrote:
All,
By George, I've got it! I reloaded -mlbeta- again, added two codes only
referring to the -nocons- options, and:
Glad you've got it going. Sounds like it was close to being right a few
messages ago, but adding noconstant to the dispersion equation zapped
you. I'm not sure what made the collinearity issue go away though. A few
comments on the code:
version 8
syntax varlist [if][in] [,DIspersion(varlist)] [,Robust] [,svy] [,Nocons]
As things now stand, the -nocons- option doesn't do anything because it is
not used in the subsequent code. I'll note below how to change that.
tokenize `varlist'
marksample touse
if "`svy'"!=""{
svymarkout `touse'
touse is not getting used in the subsequent code. Further, md cases in the
dispersion varlist aren't getting knocked out. So, this isn't coded right,
but maybe it doesn't hurt you. However, I'm guessing it would be good to
add a line like
markout `touse' `dispersion'
ml model lf mlbeta_ll (`lhs'=`rhs', noconstant) (`dispersion') `if' `in',
`robust' `svy'
What I think you want is
ml model lf mlbeta_ll (`lhs'=`rhs', `nocons') (`dispersion') if `touse',
`robust' `svy'
When calling mlbeta, it will then be your choice whether to include the
constant or not.
*
* 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/
