Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: xtmixed


From   Joseph Coveney <jcoveney@bigplanet.com>
To   Statalist <statalist@hsphsun2.harvard.edu>
Subject   Re: st: xtmixed
Date   Sat, 14 Oct 2006 06:20:52 +0900

Ali Khashan wrote:

I analysed birthweight data using OLS regression, but
because the model included gestational age the
constant variance assumption was violated. the model
included 2 categorical variables as well.

I checked the birthweight variance and found that I
can group gestational age into 5 groups each has its
own variance. I fitted the models after stratifying
gestational age and it worked but I can't get an
overall estimate.
I am now trying to use xtmixed which allows for groups
to have different variance and gives an overall
estimate as I understand.
However, I struggled to find the right options.
my model is
xi: regress bwt gest(in weeks from 23-44)
i.smoking(3categories)and i.parity(3 cat).
could any one please tell me what is the right xtmixed
code in this case. in SAS there is an option for
variance component VC. what is the equivalent in
stata?

--------------------------------------------------------------------------------

Have you considered a variance-stabilizing transformation?  Or
heteroscedastic regression (-help regh-)?  Or a generalized linear model
with a suitable link function (-help glm-)?  Or, given that you've already
categorized the continuous variable, even variance-weighted least-squares
regression (-help vwls-)?

Variance components just doesn't sound right here, regardless of whether SAS
or Stata.

Joseph Coveney

*
*   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/



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