Statalist The Stata Listserver


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

st: gllamm


From   w.mueller@uvt.nl
To   statalist@hsphsun2.harvard.edu
Subject   st: gllamm
Date   Wed, 30 May 2007 21:23:44 +0200

Hi,

Could someone please tell me what the exact econometric model for the
following two gllamm estimations commands is?

1.
gllamm var1 var2, i(person group) family(binomial) link(probit)

Here, "var1" is a binary variable, "person" is a subject identifier and
"group" is the group identifier to which a subject belongs.


2.
gllamm var3 var4, i(person group) family(gaussian) link(id)

Here var3 is a continuous variable and "person" and "group" are as in 1.
above.


According to my reading of the Skrondal/Rabe-Hesketh book, the model for
1 should be:

var1_{ijk}=β₀ + β₁*var2 + β₂*person_{ijk} + η_{jk}⁽²⁾ + η_{k}⁽³⁾ + ζ_{j} + ε_{ij}

where person_{ijk} are dummy variables for subjects, η_{jk}⁽²⁾ is the
random intercept for subject j in group k, η_{k}⁽³⁾ is the random
intercept for group k, ζ_{j} is a time-constant error component which
varies between subjects and ε_{ij} is a transitory error component which
varies over occasions i and subjects j. The random intercepts are
assumed to be independently normally distributed. Is this correct? What
about the model for 2.?


Best, Wieland.


*
* 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