Re: conf_map_counts and C statistics

On Mon, Apr 14, 2003 at 11:17 -0400, Marc Audard wrote:
> John-
> 
> Does the subplex convergence method work for conf_map_counts?
> At every step, ISIS complains and writes an error message:
> 
> *** Error:  function evaluation failed
> 
> Regards,
> 
> 	Marc

Hi Marc,

Yes, it should work.

If a failure occurs when the fit-function is evaluated, it is
usually due to the fit-parameters being out of range. Are
min/max ranges provided for all variable fit parameters?

In mapping out 2D confidence contours, the mapped parameters
may be frozen at values far from their best fit values.  If
other fit parameters are unconstrained, it may happen that, in
searching for the best fit under these condidtions, subplex can
end up in a region of parameter space where the function
evaluation fails.

To work around this, you might specify smaller min/max ranges
for all the parameters, to constrain the search to a more
well-defined region of parameter space.

Does that sound reasonable?

Thanks,

John

-- 
John C. Houck                       MIT Center for Space Research
NE80-6005: 617-253-3849                   77 Massachusetts Avenue
42:21:55.105N, 71:05:28.122W             Cambridge, MA 02139-4307
----
You received this message because you are
subscribed to the isis-users list.
To unsubscribe, send a message to
isis-users-request_at_email.domain.hiddenwith the first line of the message as:
unsubscribe