My suggestion is don't use gllamm to fit SEM.
1. It is very slow.
2. It doesn't produce any of the fit indices, although you may possibly
work some of them out using the likelihood.
3. It doesn't tell you how to modify your model.
4. Possibly it doesn't allow you to have certain correlation pattern,
although I'm not sure.
For more information, you may want to subscribe to SEMNET, a discussion
list dedicated to SEM.
Tim
Nick Cox <[email protected]>
05/12/2006 11:30
Please respond to
[email protected]
To
[email protected]
cc
Subject
RE: re: st: RE: fit SEM using gllamm
Someone may be willing to do this, but in broad
terms very open-ended questions like this are
very difficult to answer and usually ignored.
And your reply is still sent as HTML/MIME.
Please respect the list by learning how to
post to it in the ways requested.
Nick
[email protected]
Lian,Qiguo
Dear Nick Cox,
thank you anyway, but i cannot found the information i need in GLLAMM
manual.
That's why i came here for help.
The gllamm syntax is so complex beyond my understanding, i hope someone
here can give me more detail information(dataset and do file for practice
are appreciated).
I just want to fit SEM and donnot want to use other packages like AMOS.
Nick Cox
Sorry, but very general questions like this are unanswerable.
There really is no substitute for reading the documentation.
-findit gllamm- and follow the links.
(Please don't post HTML/MIME to the list. This in turn
is well documented.)
Lian, Qiguo
Can anyone be kindly help me to fit SEM using gllamm?
*
* 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/
