Re: Goodness of fit test for phase-type distribution
From: Herman Rubin (hrubin_at_odds.stat.purdue.edu)
Date: 07/29/04
- Next message: Evgenij Barsukov: "Re: GOD=G_uv BAYESIAN P=99.9 %"
- Previous message: The Ghost In The Machine: "Re: GOD=G_uv BAYESIAN P=99.9 %"
- In reply to: Arike A.: "Goodness of fit test for phase-type distribution"
- Messages sorted by: [ date ] [ thread ]
Date: 29 Jul 2004 13:33:42 -0500
In article <rieab52unevv@legacy>, Arike A. <adexby@yahoo.com> wrote:
>Hello,
>I wonder if there is any existing statistical software for
>implementing K-S and A-D goodness of fit test for fitted phase-type
>distribution. If there is none, how do I go about implementing these
>goodness of fit tests on fitted phase-type distribution.
>Thanks.
The K-S and A-D tests are not distribution free for
fitted parameters, and there are few cases in which
even the asymptotic distribution can be calculated
in closed form at the present times; most common
situations, as normal, do not belong to that class.
Unless it is one of the few which have been done,
you will have to use simulation on the distribution.
There are things which can be done to reduce the
work to simulate the tails, but these are also not
simple to describe.
-- 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@stat.purdue.edu Phone: (765)494-6054 FAX: (765)494-0558
- Next message: Evgenij Barsukov: "Re: GOD=G_uv BAYESIAN P=99.9 %"
- Previous message: The Ghost In The Machine: "Re: GOD=G_uv BAYESIAN P=99.9 %"
- In reply to: Arike A.: "Goodness of fit test for phase-type distribution"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
|
