I am running an arch-in-mean estimation involving an enormous varcov matrix.
xi: arch rva compadv i.ctry i.isic, bhhh archm arch(1) constraints(1) het(code1-code414)
where constraint define 1 [ARCH]_b[L.arch] = 0
Everything runs smoothly, except that - as often with arch - convergence takes a million years.
In more than one cases, I actually end up having "backed up" likelihood, over and over again, and no convergence.
Is there a way of extracting the intermediate coefficient estimates that correspond to a (non converged) likelihood value?
*
* 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/
