> It would be nice if someone could do a Stata package for polychoric
> correlations, which of course would include the tetrachoric case. The
> tetrac package, by its own admission, is not very accurate. There are
> some implementations out there (SPSS, Pascal, Fortran) that could be
> translated, with authors' permissions, of course.
I would need to write a help file for my ado file, and then it would be at
least the first approximation to a polychoric suite. It uses standard
Stata's -ml-... and the inaccuracy I can admit is that it implements a
two-step procedure of first determining the thresholds from univariate
data and then uses those thresholds in estimation of the correlation
itself. It does not lead to asymptotically normal and efficient estimates,
but the only paper I've seen on comparison between the two said they did
not find any practical differences -- only something in fifth the digit or
so.
--- 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/
