Mail Archive Search: <+from:dmck@silvae.cfr.washington.edu>

Results:

References: [ +from:dmck@silvae.cfr.washington.edu: 13 ]

Total 13 documents matching your query.

1. Re: Summary: bug-let in predict.glm() (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Wed, 6 Dec 2000 10:42:11 -0800 (PST); Doesn't this make sense though? On what basis could one predict a response at an unmodelled level of a factor? Thanks for enlightenment.. _____________________________________________________________; /archives/html/s-news/2000-12/msg00055.html (8,017 bytes)
2. RE: Data analysis of Florida, USA voting patterns (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Thu, 9 Nov 2000 23:34:07 -0800 (PST); There is a disturbing long-term (at least 30 years) trend in American dictionaries to change preferred meanings and pronunciations if enough people get them wrong for long enough. Some argue that thi; /archives/html/s-news/2000-11/msg00097.html (11,532 bytes)
3. coerce a list to a character vector (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Fri, 27 Oct 2000 11:23:37 -0700 (PDT); Is it possible to transfom a list with components of uneven length, but the same data type, say character, into a single vector? For example, I would like $a1: [1] "A" "N" "C" "Q" $b1: [1] "B" "A" "X; /archives/html/s-news/2000-10/msg00231.html (7,354 bytes)
4. coerce a list to a character vector (fwd) (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Fri, 27 Oct 2000 11:33:45 -0700 (PDT); See below. unlist() does it. Thanks to Alan and Rolf. _______________________________________________________________________ DON MCKENZIE Research Ecologist College of Forest Resources, Box 352100 U; /archives/html/s-news/2000-10/msg00232.html (8,387 bytes)
5. [S] S to R and back (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Sun, 23 Jul 2000 21:22:34 -0700 (PDT); Having just installed R on my linux machine (running redhat 6.0), I naively went to port datasets (that had been exported with data.dump()) from Splus 2000 for Windows to the R environment. Of course; /archives/html/s-news/2000-07/msg00179.html (7,991 bytes)
6. [S] linux vs NT performance bottlenecks (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Tue, 23 May 2000 15:02:57 -0700 (PDT); s t; /archives/html/s-news/2000-05/msg00205.html (8,784 bytes)
7. Re: [S] Coefficients of the GLM (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Wed, 16 Feb 2000 15:36:38 -0800 (PST); Is there a function available that does for glms what dummy.coef() does for anova models? This would produce the output Yimin is seeking, if I understand him correctly. Thanks. Don -- This message wa; /archives/html/s-news/2000-02/msg00138.html (13,080 bytes)
8. Re: [S] New S+2000 Patch 2 (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Wed, 10 Nov 1999 08:47:06 -0800 (PST); Amen! I did this also. Couldn't Mathsoft have handled this a bit more _______________________________________________________________________ DON MCKENZIE Research Ecologist College of Forest Resourc; /archives/html/s-news/1999-11/msg00106.html (10,197 bytes)
9. Re: [S] Canonical Correlation Analysis (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Tue, 19 Oct 1999 12:19:32 -0700 (PDT); The CCA algorithm requires you to invert the covariance matrices of X and Y. This will (should) bomb in any software package because if p,q, > n the matrices are singular. Methods like CCA and factor; /archives/html/s-news/1999-10/msg00165.html (10,500 bytes)
10. [S] error optimism in bootstrapped regression (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Tue, 12 Jan 1999 11:44:26 -0800 (PST); Please excuse that this is not directly related to Splus use. I am trying to obtain "refined bootstrap estimates" of prediction error in regression models, using the procedure given in: Efron,B. and; /archives/html/s-news/1999-01/msg00062.html (8,629 bytes)
11. [S] Variable names in plot of regression tree (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Wed, 4 Nov 1998 16:47:42 -0800 (PST); Dear Snews folks: I would like to change the names of the element of a tree object that are plotted with text.tree(). I havetried changing all the attributes of mytree$terms in vain (e.g. term.labels; /archives/html/s-news/1998-11/msg00017.html (7,747 bytes)
12. [S] variable names on regression tree plot (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Thu, 5 Nov 1998 13:36:02 -0800 (PST); Thanks to Brian Ripley, Michael Bramley, and Sam Buttrey for prompt replies. My question: Dr. Ripley pointed out that the names are as.character(mytree$frame$var) in mytree$frame, and is clear in the; /archives/html/s-news/1998-11/msg00022.html (8,348 bytes)
13. [S] modifying text.tree() (score: 1): Author: Don McKenzie <dmck@silvae.cfr.washington.edu>; Date: Thu, 23 Jul 1998 12:30:45 -0700; I would like to create a modified version of text.tree() to write back-transformed fitted values at the terminal nodes. For example: I am modeling proportional data using an arcsin/sqrt transformati; /archives/html/s-news/1998-07/msg00165.html (10,646 bytes)

Current List: 1 - 13
Page: [1]