Maarten is correct: Nick actually reported that he is using a different
(binary)dependent variable. I wrongly assumed that Nick eventually decided
to consider the child_123 var as the dependent one.
Hence, following Maarten's suggestion would solve Nick's problem.
Please apologize for my previous misreading.
Kind Regards,
Carlo
-----Messaggio originale-----
Da: [email protected]
[mailto:[email protected]] Per conto di Maarten buis
Inviato: lunedì 23 novembre 2009 12.44
A: [email protected]
Oggetto: Re: R: st: combine indicator values
--- On Mon, 23/11/09, Carlo Lazzaro wrote:
> Nick is seemingly drifting away from his original query.
> If Nick plugs in the model the child_123 var as he
> reported, he seems to leave the logistic regression
> framework in favour of the Poisson one (that is, he
> replaces binary response with count data).
No, he reported that he is using a different (binary)
dependent variable. So his problem seems to be how to
add a categorical variable to a logistic regression
model (with or without -xi-). He wants to create a
categorical variable from a set of dummy variables,
and than use -xi- to put them (as a set of dummies)
in logistic regression. He is just not sure how
to set the reference category when adding a set
of dummy variables directly rather than through
-xi-, which I explained in the posts befor.
Hope this helps,
Maarten
--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://www.maartenbuis.nl
--------------------------
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
