Alita Nandi <[email protected]> is getting an error message from
-ml check- when -ml max- would otherwise produce estimation results:
> --ml check-- gives the error message "cannot find feasible values" when it
> is testing to see whether the likelihood evaluator (the program that I have
> written to calculate the log-likelihood) works or not. But when I run --ml
> maximize-- without running --ml check-- before it, the maximization works
> and I get the estimates. By using --ml search-- before --ml maximize-- I
> get the same estimates. What I want to know is whether it is possible for
> --ml check-- to show that there are problems with my likelihood evaluator
> even if there is none. If not, what kind of errors in my programming (or
> perhaps the data set) may cause this discrepancy? I have not written the
> program code in this email so as not to clutter it. But please let me know
> if it is necessary to see the program code in order to be able to provide
> answers to my questions and I will do so.
>
> I am running Intercooled STATA Version 8 on Windows.
We are confident that the results from -ml max- are reasonable; however, we
are unable to reproduce the problem with -ml check- that Alita mentions.
If Alita will send me the likelihood evaluator and a dataset with a log that
show the steps that result in the error message, we will look into this and
report on what we find.
--Jeff
[email protected]
*
* 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/
