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, is already up and running.

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

Re: st: is gllamm appropriate? is it necessary?-more information

From   Maarten buis <>
Subject   Re: st: is gllamm appropriate? is it necessary?-more information
Date   Wed, 17 Mar 2010 08:10:42 -0700 (PDT)

--- On Wed, 17/3/10, Jessica Bishop-Royse wrote:
> 3.  As of now, my cause of death variable is 10
> categories, 9 causes and survival.
> . gllamm causeofdeath black biryear, i(countynumber)

If causeofdeath is categorical than at least you should specify the 
-link()- and -family()- options as now you modeling the "mean cause
of death" and since the coding of such categorical variables is
arbitrary the resulting means are meaningless. 

One thing you can try to make -gllamm- converge (and make the results
easier to interpret, especially with interactions) is to center the
variables, or at least make sure that each variable is 0 at some 
meaningfull value within the range of your variable , e.g. substract 
50 from age to make sure that the constant refers to 50 year olds
rather than newly born babies. Year of birth is absolutely notorious,
if you do not center the variable, the constant will refer to the 
year 0, which is usually way way way outside the range of the data.

Hope this helps,

Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen


*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index