Hi Clive,
The following seems to work:
webuse union, clear
capture program drop gplogit
program define gplogit
version 6
args lnf theta delta
qui replace `lnf' = $ML_y1*`theta'*(1+`delta') - ///
ln(1+exp(`theta'*(1+`delta')))
end
ml model lf gplogit (union = age black grade south) ///
(delta: black,nocons)
ml maximize
Note that changing the "qui replace `lnf'..." line to:
qui replace `lnf' = $ML_y1*`theta'*(exp(`delta')) - ///
ln(1+exp(`theta'*(exp(`delta'))))
forces the estimated scale parameter/ error variance to be positive,
which is a good thing. This is a similar parameterisation to that used
in Richard Williams' -oglm- routine.
Hope this helps.
Arne
On 20/02/07, Clive Nicholas <[email protected]> wrote:
For fun, I thought I'd test out an old piece of Stata code I found in Allison (1999: 203-4) that fires up a logit model which takes into account unequal residual variation across groups. I'm sure it's just a minor thing, but I'm having problems running his Stata 6 routine, which utilizes -ml- code:
webuse union, clear
program define gplogit
version 6
args lnf theta delta
qui replace `lnf' =
$ML_y1*`theta'*(1+`delta')
ln(1+exp(`theta'*(1+`delta')))
end
ml model lf gplogit (union = age black grade south) ///
(delta: black,nocons)
ml maximize
Using Glenn Hoetker's -complogit- routine shows there is residual variation and actions the above model. Unfortunately, all I get with this code is
invalid syntax
r(198);
which is rather puzzling, as I can't see a lot wrong. Can anyone solve this puzzle?
Clive Nicholas
Reference
Allison PD (1999) "Comparing Logit and Probit Coefficients Across Groups", SOCIOLOGICAL METHODS AND RESEARCH (28)2: 186-208.
___________________________________________________________
All New Yahoo! Mail – Tired of unwanted email come-ons? Let our SpamGuard protect you. http://uk.docs.yahoo.com/nowyoucan.html
*
* 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/
*
* 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/
