home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gossip.pyramid.com!olivea!sgigate!sgiblab!darwin.sura.net!jvnc.net!netnews.upenn.edu!netnews.noc.drexel.edu!coe.drexel.edu!n1-24-215-coe-abbotts.coe.drexel.edu!chaas
- From: chaas@coe.drexel.edu (Chuck Haas)
- Newsgroups: sci.math.stat
- Subject: Goodness of Fit with Zero Degrees of Freedom
- Message-ID: <1992Oct8.144919.21524@cbis.ece.drexel.edu>
- Date: 8 Oct 92 14:49:19 GMT
- Sender: news@cbis.ece.drexel.edu
- Organization: Drexel University
- Lines: 23
- X-Xxmessage-Id: <A6F9C2718C0118D7@n1-24-215-coe-abbotts.coe.drexel.edu>
- X-Xxdate: Thu, 8 Oct 92 02:48:17 GMT
- X-Useragent: Nuntius v1.1.1d12
-
-
- Suppose I have a set of categorical data.
- trait present trait absent group totals
- group a a1 a0 at
- group b b1 b0 bt
- ... etc.
-
- The group totals should be regarded as fixed. I fit a model to this
- data. If the model has k parameters and there are m groups, then I know
- that I could do a goodness of fit test (e.g., Fisher chi-square or
- likelihood ratio) and compare to the chi-square distribution with m-k
- degrees of freedom.
-
- Suppose, however, that I fit a model with k=m parameters. How do I then
- test goodness of fit. In other words, how much slippage should be
- allowed before I reject the null hypothesis of model adequacy
- (conditional on the parameters)?
-
- Charles N. Haas
- Professor, Environmental Engineering
- Drexel University, Philadelphia PA 19104
- ======================================
- CHaas@COE.Drexel.Edu
-