No. The only diagnostics you could have from -ml- itself is that the
likelihood is not concave at the last iteration. -ml- is way too
general to look into those issues; they are up to the researcher to
determine, and code properly. Besides, there are models that are
underidentified only for certain parameter combinations (and
empirically underidentified near them), such as y=b_0 + b_1 x^b_2
which is underID when b_1==0 (value of b_2 cannot be determined) or
b_2 == 0 (only b_0+b_1 is identified).
On 7/31/06, Lian Jian <[email protected]> wrote:
Hi folks,
I am wondering if STATA offers any diagnostics of identification for
maximum likelihood estimations (with system of equations)? I have
looked both the statalist archives and Gould, Pitblado, and Sribney's
book, maximum likelihood estimation with stata (third ed). I 'd
highly appreciate any suggestions/help.
thanks,
Lian
*
* 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/
--
Stas Kolenikov
http://stas.kolenikov.name
*
* 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/
