Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: mi estimate using gllamm and pweight


From   "Wike, Traci" <twike@email.unc.edu>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: mi estimate using gllamm and pweight
Date   Wed, 1 Jun 2011 21:45:30 +0000

Thank you all for your feedback. I have one additional question re: this topic. 

I have created my propensity score weights using logistic regression. The variable name is psb. However, when I add pweight (psb) to my model, it does not read the probability weights. How do I create these so that gllamm will use my weights?

Thanks,
traci

________________________________________
From: owner-statalist@hsphsun2.harvard.edu [owner-statalist@hsphsun2.harvard.edu] on behalf of Maarten Buis [maartenlbuis@gmail.com]
Sent: Tuesday, May 31, 2011 9:42 AM
To: statalist@hsphsun2.harvard.edu
Subject: Re: st: mi estimate using gllamm and pweight

On Tue, May 31, 2011 at 3:30 PM, Wike, Traci wrote:
> I am running a multi-level model with multiply imputed data and using propensity score weights. Because xtmixed does not allow pweights, I am using gllamm. Yet, when I try to run the model using mi estimate: (to account for the imputed data), I receive a message saying that mi estimate is not supported with gllamm. Does anyone have suggestions for how to handle this?

So you are trying to combine propensity scores with a multilevel model
with multiple imputation? If I were in a positive mood, I would say
that that is very brave. I am not, so I will say that I would not try
to combine that many techniques. Who knows what the properties of such
a combined estimator are? Each of them are reasonable approximations
of the thing they want to measure, but that does not mean that the
same is true for the combined estimator. At the very least you are
collecting an awful lot of assumptions with such a combined model.

Hope this is not too depressing,
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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index