Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: cross-classified random effects

From   "Glance, Laurent" <>
To   "''" <>
Subject   st: RE: cross-classified random effects
Date   Tue, 23 Mar 2004 07:05:19 -0500

I assume that your outcome variable is binary.
Gllaam will not handle cross-classified data structures. MLWIN can handle
cross-classified models, but the syntax for cross-classified models is not
well documented.  I have found PROC GLIMMIX to work well with
cross-classified data structures.  The best reference for PROC GLIMMIX is
SAS System for Mixed Models by Ramon Littell - available through the SAS web


-----Original Message-----
From: Do, Phoenix [] 
Sent: Monday, March 22, 2004 6:42 PM
Subject: st: cross-classified random effects


I am trying to model risk behaviors for IDUs. Since my data is hierarchical
in nature, I want to apply a multilevel model using gllamm. 

My dataset consists of an unbalanced panel in which some 
people are observed only once while others are observed 2 to 
10 times. We want to account for "neighborhood" 
characteristics so we have neighborhood level variables in 
our model. So level 1 would be the individual level time 
series data. Level 2 would be person level. Level 3 would be 
the neighborhood.

However, these people move and are not necessarily in the 
same neighborhood throughout our study.

I believe what I need to do is use a crossed random effects 

Can you do this in gllaam? And if so, how? I haven't been able to find any
reference to this in the manual.

Thank you for any help you can offer,

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index