Re: error in ml stat

On Wed, Dec 12, 2012 at 11:30 -0500, David P. Huenemoerder wrote:
> 
> We have found an error in the maximum likelihood statistic in ISIS,
> "ml".
> 
> Formula 7.51 (in the version of the manual dated 6/2/11) is incorrect.
> Instead of ln Gamma[Mi+1], it should be ln Gamma[Ci+1]. It should also
> have factor of 2 to make it distributed as chisq.  (Thanks to Herman
> Marshall for providing details.  And for being the first user,
> apparently, of "ml".)
> 
[...]

The bug in 'ml' will be fixed and the documentation corrected
in the next isis release, version number 1.6.2-14.

Thanks,
-John
----
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