Re: Anderson-darling test for discrete distribution fitting

In article <f1acnb$21s$1@xxxxxxxx>, Samik R. <samik@xxxxxxxxxxxxx> wrote:
On 5/2/2007 3:47 AM, David Jones wrote:
Samik R. wrote:
On 5/1/2007 12:06 PM, Herman Rubin wrote:
In article <f17sbo$4oa$1@xxxxxxxx>, Samik R. <samik@xxxxxxxxxxxxx>
wrote:

Not only the AD test, but all nonparametric tests,
are conservative when used to test a specific
discrete distribution. If the distribution is
not "too discrete", it might not make too much
of a difference.

As for testing a distribution with fitted parameters,
the distribution of the test statistic is likely to
be FAR off what one gets for a specific distribution.
It may not even be in the same "ballpark".
--
This address is for information only. I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Department of Statistics, Purdue University
hrubin@xxxxxxxxxxxxxxx Phone: (765)494-6054 FAX: (765)494-0558
.

Relevant Pages

  • Re: Lilliefors Test : 40 years
    ... The Lilliefors Test tests whether a distribution is ... and unknown mean, known mean and unknown variance, ... Herman Rubin, Department of Statistics, Purdue University ...
    (sci.stat.math)
  • Re: help: testing for unimodality in multiple dimensions?
    ... >continuous distribution. ... are those of the Statistics Department or of Purdue University. ... Herman Rubin, Department of Statistics, Purdue University ...
    (sci.stat.consult)
  • Re: Kolmogorov-Smirnov test... a good overview
    ... the distribution of the KS ... of statistics is "self-thought" with applications in mind. ... are those of the Statistics Department or of Purdue University. ... Herman Rubin, Department of Statistics, Purdue University ...
    (sci.stat.math)
  • Re: Does linearly dependent imply statistically dependent?
    ... >If X, Y, and Z have non-degenerate continuous distribution, and they ... If X, Y, and Z has a non-planar distribution, ... Herman Rubin, Department of Statistics, Purdue University ...
    (sci.math)
  • Re: Anderson-darling test for discrete distribution fitting
    ... On 5/2/2007 11:00 AM, David Jones wrote: ... have a reasonable formulation for the AD statistic, ... the AD test as usually done, is essentially a test of whether certain transformed values of the data have a uniform distribution ... ... For a discrete distribution you might want the target to be 1/N for the lowest observation, ...
    (sci.stat.consult)