Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: problem with convergence


From   Woolton Lee <finished07@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: problem with convergence
Date   Thu, 17 Jun 2010 13:06:33 -0400

I am estimating models separately on several markets each with several
thousand observations.  To calculate standard errors for certain
calculations that I need to do, I am bootstrapping the sample (using a
sample size equal to the original sample), but I am finding that for a
small subset of the markets the model I am using does not converge in
the bootstrapped sample (I am using only 100 bootstraps).  This seems
to occur only sporadically.  For example, of the 100 bootstraps non
convergence occurred on the 68th bootstrap in one market and the 43rd
in another market.  So I want to include a line that counts the number
of iterations a model has run through for the log likelihood and have
it redraw the bootstrap sample if convergence does not occur.

Is there any way to do this?  I am using a user written estimation
command called multin which uses ml but I am not certain it there is a
simple way to do this.

Thank you for your help,

Woolton
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index