Re: Behavior for conf() different from XSPEC (bug?)

[Apologies for the duplicate post -- I wanted to re-submit
this with the correct subject line so that my reply wouldn't
get lost - JH]

On Mon, Jun 07, 2004 at 09:44 -0400, nrbutler_at_email.domain.hidden> 
> Hey,
> 
> I haven't seen anyone comment on this before.  Apologies if that's not the 
> case.
> 
> I've noticed that the conf() function in ISIS (and fconf() and vconf())
> determine parameter confidence intervals by freezing the other parameters
> and by varying the interesting parameter until a desired dchi^2 is
> achieved.  

That's not the way it works -- what makes you think so?
You can use vconf to see how all non-frozen parameters 
are varied during the confidence limit search.


>            I'm not sure what this means statistically. 
> 
> The standard XSPEC approach is to fit the other (non-interesting) parameters
> as the interesting parameter is varied.  When the desired dchi^2 is found
> in this case, the parameter bounds are confidence regions based on the
> likelihood ratio test.  This is valid if your number of data points and counts
> are large.

ISIS does it the same way.

> 
> In practice, the important thing I've noticed is that ISIS underestimates
> confidence intervals relative to XSPEC.  If this isn't a bug, it would
> be a good idea for there to be more explanation in the manual and help.
> 
> -Nat

Can you quantify this?  If there's a significant consistent
difference, I'd like to understand it.

Thanks,
-John

-- 
John C. Houck         MIT Center for Space Research
tel: 617-253-3849           77 Massachusetts Avenue
fax: 617-253-8084          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