Chia C Chong <[EMAIL PROTECTED]> wrote in message
a8s4t4$j7d$[EMAIL PROTECTED]">news:a8s4t4$j7d$[EMAIL PROTECTED]...
>
> "Glen Barnett" <[EMAIL PROTECTED]> wrote in message
> a8qu0t$ri6$[EMAIL PROTECTED]">news:a8qu0t$ri6$[EMAIL PROTECTED]...
> >
> > Chia C Chong <[EMAIL PROTECTED]> wrote in message
> > news:a8qko6$ftk$[EMAIL PROTECTED]...
> > > I would like to get more infor for the above topic. Anyone know any good
> > > sources for that? I will have to implement the above test later on for
> to
> > > perform some goodness-of-fit test on my data.
> >
> > Why will you /have/ to implement a Cram�r-von Mises test?
>
> I would like to test whether the theoretical distribution I used to
> represent the empirical data is reasonable....

Dealing with that question may not need a hypothesis test at all, but even if
that were
the case, again, why does it /have/ (in your words) to be C-vM and not
something else?

There are other tests. Anderson-Darling is more readily available in packages,
for example.

Glen



.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to