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: generating predicted values for growth models


From   Amy Hsin <hsin.amy@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: generating predicted values for growth models
Date   Fri, 29 Jun 2012 16:53:16 +0200

H Maarten,

Thanks! That was helpful.  However, I still encounter the same problem
when I revise the code.

Now the command is the following.  However, the predicted values
remain unchanged (i.e. they are the same as the observed).  Any
suggestions?

xtmixed gpach i.wave##i.asian i.wave##i.female i.wave##c.psescomk8 if
gpachmiss!=1 || id: wave , cov(unstructured) variance emiterate(250)
mle;

margins, over(asian wave);


Thank you,
Amy

On Fri, Jun 29, 2012 at 4:23 PM, Maarten Buis <maartenlbuis@gmail.com> wrote:
> One thing you did wrong is that you used the -xi:- prefix and the
> -margins- post-estimation command. -margins- needs to know about the
> interactions you created, and -xi:- will does not leave that
> information behind. So you need to use the factor variable notation
> instead, see -help fvvarlist-. Translating your command would be:
>
>  xtmixed gpach i.wave##(i.race i.female c.psescomk8) <other stuff>
>
> -- Maarten
>
> On Fri, Jun 29, 2012 at 4:03 PM, Amy Hsin wrote:
>> Hi,
>>
>> I am trying to generate predicted values from linear growth curves and
>> am running into problems. The predicted values that I am generating
>> using models that adjusts for a variety of covariates are the same as
>> the observed values, which does not make sense. Is there something
>> that I am obviously doing wrong?
>>
>> Here is a simplified version of my model.
>>
>> xi: xtmixed gpach i.wave*race i.wave*female i.wave*psescomk8 if
>> gpachmiss!=1 || id: wave , cov(unstructured) variance emiterate(250)
>> mle;
>> margins, over(race wave);
>>
>> -gpach are test scores evaluated at time 0, 1, 2 and 3
>> -race and female are both dichotomous variables
>> -psescomk8 a time invariant continuous covariate
>> In this model, I'd like to measure time nonparametrically so I am
>> including it as a series of dummy variables.
>> When I use the "margins" command to estimate predicted test scores for
>> each racial category at each point in time, I get values that are
>> identical to the observed values.
>>
>> Any ideas as to what I am doing wrong?
>>
>> Thank you in advance.
>>
>>
>> Amy Hsin
>> *
>> *   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/
>
>
>
> --
> --------------------------
> 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/



-- 
Amy Hsin
Assistant Professor of Sociology
Queens College, City University of New York
http://qcpages.qc.cuny.edu/~ahsin

*
*   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