home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!pipex!warwick!uknet!comlab.ox.ac.uk!ph
- From: ph@physiology.oxford.ac.uk (Patrick Haggard)
- Newsgroups: bit.listserv.stat-l
- Subject: terminology of proc glm; manova /canonical; output
- Message-ID: <1993Jan21.183847.25291@galen.physiol.ox.ac.uk>
- Date: 21 Jan 93 18:38:47 GMT
- Sender: ph@physiol.ox.ac.uk (Patrick Haggard)
- Organization: Physiology Department, Oxford University, Oxford, UK.
- Lines: 25
- Originator: ph@galen.physiol
-
- Hello. I've been using proc glm manova to compare data in two
- conditions, and I want to look at the differences between the
- conditions in terms of a canonical variate which optimally
- discriminates between the two groups. manova /canonical outputs
- five sets of canonical variates, (total, between, within, standardized
- and raw), and I'm not familiar with the terminology. Which is the
- actual canonical discriminant function, what is the standardized
- ne standardized to, and what is the meaning of the total one? My
- books on multivariate analysis seem to suggest there's just one,
- not five!
-
- Guidance much appreciated
-
- Patrick
-
- P.S. Is it true that the SAS tech notes can be got by ftp?
-
- --------------------------------------------------------------------------
- Patrick Haggard Email (WORLD): ph@physiol.ox.ac.uk
- University Laboratory of Physiology Email (JANET): ph@uk.ac.ox.physiol
- Parks Road, Tel. (0865) 272116
- Oxford, OX1 3PT Fax. (0865) 272469
- England
- --------------------------------------------------------------------------
-