On Fri, 2007-09-03 at 20:41 +0100, Eike Rathke wrote:
> On Friday, 2007-03-09 19:32:01 +0100, Eike Rathke wrote:
>
> > Furthermore, Ecma/Excel's definition of degrees of freedom looks weird,
> > for rows>1 and cols>1 they say it would be (rows-1)*(cols-1). Why?!?
>
> Btw, OOo does the same. Probably because it was documented as such.
Which test is CHITEST supposed to perform?
In the cases of a test for independence or a test of homogeneity, these
degrees of freedom make sense. (But we should have only one regular
region).
In the case of a Goodness-of-Fit test one would have observed and
expected frequencies (corresponding to each other) and the degree of
freedom should be n-1.
They all use the chi-square distribution and are referreed to as
chi-tests.
Andreas
--
Andreas J. Guelzow, Professor
Dept. of Mathematical & Computing Sciences
Concordia University College of Alberta
