On 2004-11-10, at 16.00, Guillaume Frechette wrote:
Dear gllamm users, I am doing a simulation using gllamm and I want to
limit
the maximum number of iterations gllamm can do (using it(#)). After
each run
of gllamm the program saves the p-value one of the coefficients.
However,
when gllamm reaches the value specified in it(#) I would like to set
that
value to missing. Can someone tell me if gllamm saves a result that
would
allow me to write a code that sets that value to missing? Something
like the
number of iterations it performed. Thanks in advance.
The scalar -e(ic)- seems to report the number of Newton-Raphson
iterations and the log for the 20 last iterations seems to be reported
in the matrix -e(ilog)-. If I remember correctly from the manuals and
the book on ml estimation by Gould et al, these results are posted
directly by -ml-.
However, there seem to be no report of the adaptive quadrature
iterations which complicate things i.e. if you want to use the -adapt-
option (as you probably should).
Michael
*
* 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/
