Does anyone have experience with Stata ML?  My ML program runs with one data set
but with a second data set I get the error message ?could not find feasible
values?.  It gives this error even if I initialize with the actual
convergence values and even if I specify ?difficult.?

In principle the model can be fitted because the data is from a paper by Imbens
that describes the method.  (I?m trying to replicate the Imbens result to
make sure my program works correctly.)

Should I: (1) program the maximization routine myself? (2) try to write a
?d2? program instead of the ?lf? program so that I can specify the
second derivatives?  (3) keep debugging?!

Does anyone know if Mata might be the solution?  Any suggestions?

Thanks for your help.
Joel Middleton

**********************************************************
             Political Methodology E-Mail List
   Editors: Melanie Goodrich, <[log in to unmask]>
            Delia Bailey, <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************