There is a von Mises distribution on the circle.
Cramer had nothing to do with that.
The Cramer-von Mises test is, as implied earlier,
a relative of a test applied to circular data.
Nick
[email protected]
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of
> FEIVESON, ALAN
> H. (AL) (JSC-SK) (NASA)
> Sent: 15 January 2004 21:47
> To: '[email protected]'
> Subject: st: RE: RE: RE: Query: The Cramer-von Mises Test
>
>
> Amani -
>
> The Cramer- von Mises (CVM) test is for goodness of fit -
> not to be confused
> with the CVM distribution on a circle. I believe the CVM
> test statistic is
> something like the integral of the squared difference between the
> hypothetical and empirical probability densities (as opposed to the
> Kolmorogov Smirnov statistic which compares the the CDFs using an
> absolute-value metric.
>
> Al Feiveson
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of
> [email protected]
> Sent: Thursday, January 15, 2004 11:18 AM
> To: [email protected]
> Subject: st: RE: RE: Query: The Cramer-von Mises Test
>
>
> My thanks to your reply Nick, I did see a pointer to your
> -circsumm- using
> "net search Cramer-von Mises" but I did not immediately
> relate that to the
> paper.
>
> I will have a look at circsumm.ado and come back to you for
> more advice.
>
> Best regards,
>
> Amani
>
> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Nick Cox
> Sent: Wednesday, 14 January 2004 16:12
> To: [email protected]
> Subject: st: RE: Query: The Cramer-von Mises Test
>
>
> I have used material from this reference in my program
> -circsumm- which is available directly from SSC.
>
> The specific test I used was the Kuiper test
> for circular data. I guess this is related to, but
> not identical to, what you want.
>
> The code -circsumm.ado- may therefore be a stepping
> stone to what you want.
>
> Just yesterday I updated the program to -circsummarize-.
> That is not yet on SSC.
>
> Nick
> [email protected]
>
> [email protected]
> >
> > I need to use the Cramer-von Mises test for some
> > comparisons (ref: Stephens,
> > M. A. (1970). Use of the Kolmogorov-Smirnov, Cramer-von
> > Mises and Related
> > Statistics Without Extensive Tables. Journal of the Royal
> > Statistical
> > Society, Series B, 32, 115-122).
> >
> > I was wondering whether someone has written an ado for that
> > - I have already
> > followed the regular stata searches (by "findit" and
> > looking through the
> > Statalist archives) and there seems to be nothing on offer.
>
*
* 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/
