> I know that Stata doesn't support SEM, but I am seeking the
> advice of you who had to work with SEM about which software
> to choose. I am not sure which one is better AMOS or Lisrel.
They have slightly different capabilities. Lisrel seems to be better with
dealing with polychoric correlations and thus data of discrete nature, and
AMOS is better with handling missing data and the bootstrap (although
after having taking a resampling class from Ed Carlstein who was the first
to propose block bootstrap for time series years ago, I lost all of the
trust in the method whatsoever. There are way too many assumptions one
would need to make, or the data should satisfy, to justify it). On the
interface side, Lisrel is more like Stata as it is a command prompt/script
driven, so you would need to invest some time into it, while AMOS is a
point-n-click package that is fairly easy to get started with pretty much
right away.
And yes you can program your model if it is not too complicated either
direclty through Stata's -ml-, or through -gllamm- which seems to be a
magic program to me -- I've tried to figure it out at least three times in
the last two years, but never fully succeeded. The old Stata joke "Let's
make it another option for -regress-" will soon become "Let's make it
another option for -gllamm-" :).
--- Stas Kolenikov
-- Ph.D. student in Statistics at UNC-Chapel Hill
- http://www.komkon.org/~tacik/ -- [email protected]
* This e-mail and all attachments to it are not intended to provide any
* reasonable point of view and was transmitted to you in error. It
* should be immediately deleted by all recepients unless they really
* enjoy communicating with the author :). Other restrictions apply.
*
* 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/
