I believe you can also run a log-linear model using the poisson command.
Here you can analyze all three dependent variables at once.
Maria
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Christian
Hunkler
Sent: Tuesday, March 22, 2005 10:33 AM
To: [email protected]
Subject: Re: st: Logistic regression for multiple binary dependent
maybe a multinomial logit model is appropriate for your analysis (mlogit
command in stata), if your three dependent variables are somehow similar to
e.g. marital status (married, divorced, never married, widowed). If your
three dep. vars are unrelated, or an observation has a one one two of the
three mlogit is not appropriate. In this case you need to run three single
models or you need to combine the three dependent variables to one
meaningful outcome variable.
Chris
***
Sonderforschungsbereich 504, Universti�t Mannheim
[email protected]
At 16:10 22.03.2005, you wrote:
>Hi list,
>
>In situations where we have a binary (0/1) dependent variable and it is
>well known to use logistic regression for the data analysis. However, we
>have a study with three binary dependent variables and we are looking for
>some statistical procedure to analyze them together.
>
>Any idea?
>
>Thanks in advance,
>
>/Christian
>
>
>
>
>*
>* 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/
*
* 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/
