st: weird error in -nlcom-

["Feiveson, Alan H. (JSC-SK311)" <alan.h.feiveson@nasa.gov>]

Hi - In Stata 10 (Windows), after runnning a -ml- program (of my own), I
tried to use -nlcom- .

Depending on what values I give it (see below) it either works, or gives
a peculiar error messgae " maximum number of iterations exceeded".

First of all, -nlcom- is supposed to use the delta method, which doesn't
involve iterating anything - so what's going on?

Even more mystifying is that if I slightly change the inputs, it works
or doesn't work (see below)



. run flare_norm t1 t2 868 366 1900 2100
Initial Values: t0 = 2100,  sig = 868     K = 366 

                                                  Number of obs   =
365
                                                  Wald chi2(0)    =
.
Log likelihood = -1151.2251                       Prob > chi2     =
.

------------------------------------------------------------------------
------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
lnK          |
       _cons |     5.9161   .0525831   112.51   0.000     5.813039
6.019161
-------------+----------------------------------------------------------
------
lnsig        |
       _cons |   6.703776   .0463619   144.60   0.000     6.612908
6.794644
-------------+----------------------------------------------------------
------
t0           |
       _cons |   1857.695   45.25929    41.05   0.000     1768.988
1946.402
------------------------------------------------------------------------
------


which is fine. Now I try some -nlcom- statements (see below)


. nlcom exp([lnK]_cons)*(norm( (2039-[t0]_cons)/exp([lnsig]_cons)) -
norm( (1679-[t0]_cons)/exp([lnsig]
> _cons)))/5

       _nl_1:  exp([lnK]_cons)*(norm(
(2039-[t0]_cons)/exp([lnsig]_cons)) - norm( (1679-[t0]_cons)/exp(
> [lnsig]_cons)))/5

------------------------------------------------------------------------
------
             |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------
------
       _nl_1 |   12.96116   .8592571    15.08   0.000     11.27704
14.64527
------------------------------------------------------------------------
------

This one works. Now if I change "2039" to "2038" and "1679" to "1678",
it crashes:



. nlcom exp([lnK]_cons)*(norm( (2038-[t0]_cons)/exp([lnsig]_cons)) -
norm( (1678-[t0]_cons)/exp([lnsig]
> _cons)))/5

       _nl_1:  exp([lnK]_cons)*(norm(
(2038-[t0]_cons)/exp([lnsig]_cons)) - norm( (1678-[t0]_cons)/exp(
> [lnsig]_cons)))/5
Maximum number of iterations exceeded.
r(498);


Any ideas what's happening? By the way, the same sort of thing was
occuring in Stata 9.

Al Feiveson

*
*   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/