st: ML maximize error message

[antonio martuscelli <totissoit@yahoo.it>]

Dear all,

I am trying to maximize a complex likelihood function and when I use
the dfp or bfgs techniques I get the error message below. This only
happens if I use those techniques and not if I use the newton-raphson
(in which case it is just very slow). Any idea what is the reason?

Thanks a lot.

. ml maximize, difficult noclear
initial:       log likelihood = -4341.6282
rescale:    log likelihood = -4341.6282
rescale eq:    log likelihood = -4339.4255
(setting technique to bfgs)

 *:  3200  conformability error
   opt__looputil_step():     -  function returned error
   opt__loop_cycle():        -  function returned error
           _moptimize():       -  function returned error
          Mopt_maxmin():    -  function returned error
                <istmt>:         -  function returned error
r(3200);
*
*   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/