Mail Archive Search: <+subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[S\]\s+stepwise\s+discriminant\s+analysis\s*$/>

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[S\]\s+stepwise\s+discriminant\s+analysis\s*$/: 2 ]

Total 2 documents matching your query.

1. [S] stepwise discriminant analysis (score: 1): Author: "William M. Grove" <William.M.Grove-1@tc.umn.edu>; Date: Mon, 10 Aug 1998 11:18:37 -0500; I see that step() and stepAIC() work on objects of the "lm" class (e.g., returned from glm(), gam() ), according to the documentation. As such I assume it won't work with an unaltered form of lda() o; /archives/html/s-news/1998-08/msg00057.html (9,204 bytes)
2. Re: [S] stepwise discriminant analysis (score: 1): Author: Prof Brian D Ripley <ripley@stats.ox.ac.uk>; Date: Mon, 10 Aug 1998 19:53:18 +0100 (BST); I presume that you are talking about _linear_ discriminant analysis here. And how is that useful? The point of stepwise fitting is to do worse on the training set but better on the test set. Stepwise; /archives/html/s-news/1998-08/msg00060.html (10,261 bytes)

Current List: 1 - 2
Page: [1]