I tried gllamm on a toy data set (pasted below). Rather than the variance components I hoped for, it gave the output below, including a mysterious s1 which wasnt in the data set. Am I being stupid? Are my data not set up properly? Im running Stata 8.2. Geoff
. gllamm chld, i(hh psu)
gllamm model Number of obs = 35
Model chi2(0) = .
Prob > chi2 = .
Log Likelihood = -54.0557850
------------------------------------------------------------------------------
chld | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
chld |
_cons | .720011 .2974915 2.42 0.016 .1369384 1.303083
-------------+----------------------------------------------------------------
s1 |
_cons | -.3785962 .2322276 -1.63 0.103 -.833754 .0765616
-------------+----------------------------------------------------------------
hh |
_cons | -.0934234 .1966765 -0.48 0.635 -.4789022 .2920555
-------------+----------------------------------------------------------------
psu |
_cons | -.2840285 .4295068 -0.66 0.508 -1.125846 .5577894
------------------------------------------------------------------------------
psu hh chld
1 1 -.0002322
1 2 -.8776832
1 2 .6442573
1 3 1.776474
1 3 1.69835
1 3 2.233133
1 4 -1.583588
1 5 .4658188
1 6 1.895023
1 7 -.1867006
1 7 .2097958
1 8 -.6904323
2 1 -1.346911
2 1 -.4776295
2 2 -.173507
2 3 -1.417931
2 4 .2320751
2 5 .4959524
2 5 1.034853
3 1 .334507
3 1 .3730094
3 2 .4297595
3 3 2.242641
3 4 .9147155
3 4 .8138424
4 1 .3867926
4 2 2.972212
4 2 1.865673
4 2 3.375655
4 3 .4450933
4 4 2.861456
4 5 -.3123977
4 5 1.838948
4 6 2.302191
4 7 3.418916
> -findit hierarchical random effects- brings up -gllamm-.
>
>
> Scott
>
>
> > -----Original Message-----
> > From: [email protected] [mailto:owner-
> > [email protected]] On Behalf Of Geoff Jones
> > Sent: Monday, August 08, 2005 11:13 PM
> > To: [email protected]
> > Subject: st: Random effects hierarchical models
> >
> > How can I fit a random effects model with a nested (hierarchical)
> > structure in Stata? I have measurements on children within households
> > within PSUs, and want estimates (sorry, "predictions") of the random
> > effects for households and PSUs. Quite a large data set (6000 children,
> > several hundred PSUs).
> >
> > I can do this in Splus (lme, random.effects) but need to be able to do it
> > in Stata. Any suggestions. Geoff
> >
> > *
> > * 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/
