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

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*\[S\]\s+Splus\s+mclust\s+question\s*$/: 4 ]

Total 4 documents matching your query.

1. Re: [S] Splus mclust question (score: 1): Author: Prof Brian D Ripley <ripley@stats.ox.ac.uk>; Date: Sun, 1 Oct 2000 08:09:20 +0100 (BST); mclust (like kmeans) works with an n x p matrix of n objects in p dimensions, NOT an n x n distance matrix. That sort of error usually means that you have a data frame when you think you have a matri; /archives/html/s-news/2000-10/msg00000.html (8,767 bytes)
2. Re: [S] Splus mclust question (score: 1): Author: gslaberge <gslaberge@uswest.net>; Date: Sun, 01 Oct 2000 09:34:06 -0600; Dear Professor Ripley: I now understand that mclust cannot be used with dissimilarity matricies. I have constructed my distance matrix using SAS. This distance matrix is the sum of nine euclidean dis; /archives/html/s-news/2000-10/msg00001.html (9,762 bytes)
3. Re: [S] Splus mclust question (score: 1): Author: Prof Brian D Ripley <ripley@stats.ox.ac.uk>; Date: Sun, 1 Oct 2000 16:49:00 +0100 (GMT Daylight Time); You can't. You don't have any data to test distributional properties with, as I said before. Distances are not data. You will, and the data will need to be continuous variates too. It's highly unlike; /archives/html/s-news/2000-10/msg00002.html (10,826 bytes)
4. [S] Splus mclust question (score: 1): Author: "Greggory S. LaBerge" <gslaberge@uswest.net>; Date: Sat, 30 Sep 2000 17:55:17 -0600; To members of the list: I have a n X n distance matrix from 37 human populations that summarises genetic allele frequency data. Zeros are on the diagonal and the matrix is square. I am trying to read; /archives/html/s-news/2000-09/msg00204.html (7,200 bytes)

Current List: 1 - 4
Page: [1]