Mail Archive Search: <+from:rolf@math.unb.ca>

Results:

References: [ +from:rolf@math.unb.ca: 88 ]

Total 88 documents matching your query.

1. Re: Identifying Splus vs. R (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Fri, 3 Feb 2006 10:59:31 -0400 (AST); I think if(identical(version$language,"R")) { } else { } will work, and is sufficiently sophisticated to satisfy most tastes. cheers, Rolf Turner rolf@math.unb.ca; /archives/html/s-news/2006-02/msg00011.html (6,895 bytes)
2. Re: weighted regression (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 19 Dec 2005 12:11:22 -0400 (AST); The solution to the problem is trivial **in theory**. Numerical considerations might bite you; I dunno. Let your vector of observations be Y and assume Y ~ N(X beta, Delta) where Delta = diag{sigma_1; /archives/html/s-news/2005-12/msg00059.html (9,435 bytes)
3. S and value for money. (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Wed, 9 Nov 2005 12:43:09 -0400 (AST); Use R. (a) It's generally better (e.g. plotmath()) and (b) it's free. This is not only good for you from a practical point of view, but it's good in **principle**. Software **should** be free -- in b; /archives/html/s-news/2005-11/msg00013.html (6,778 bytes)
4. Re: simple vector question (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Wed, 12 Oct 2005 13:29:58 -0300 (ADT); as.vector(outer(a,b,"+")) or c(outer(a,b,"+")); /archives/html/s-news/2005-10/msg00035.html (6,674 bytes)
5. Re: Sub setting data (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Fri, 16 Sep 2005 17:35:57 -0300 (ADT); s1 <- df1$ser.no s2 <- df2$ser.no ii <- match(s2,s1) df.new <- df1[ii,] cheers, Rolf Turner rolf@math.unb.ca Original message:; /archives/html/s-news/2005-09/msg00054.html (7,173 bytes)
6. Re: Concerns about SPLUS version 7 (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Tue, 12 Jul 2005 12:06:20 -0300 (ADT); I cannot for the life of me fathom why **anyone** would use SPLUS anymore. No one that I know does so. All use R. The only problem that ever arises is of the nature of ``Whoops! I have this data set; /archives/html/s-news/2005-07/msg00067.html (8,437 bytes)
7. Re: Concerns about SPLUS version 7 (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Tue, 12 Jul 2005 13:52:25 -0300 (ADT); I find your arguments completely unconvincing, but I am not going to waste my time debating the issue. cheers, Rolf Turner rolf@math.unb.ca; /archives/html/s-news/2005-07/msg00074.html (8,930 bytes)
8. Re: seq(0, 1, .05)==0.15 doesn't work (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Thu, 12 May 2005 08:38:34 -0300 (ADT); s; /archives/html/s-news/2005-05/msg00066.html (9,623 bytes)
9. Re: Deleting Selected Rows (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Sat, 19 Mar 2005 12:58:31 -0400 (AST); iii <- <selected rows> (e.g. iii <- c(5,17,42,55) new.df <- old.df[-iii,] cheers, Rolf Turner rolf@math.unb.ca; /archives/html/s-news/2005-03/msg00099.html (6,503 bytes)
10. Re: glm model comparison (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Tue, 3 Aug 2004 17:14:24 -0300 (ADT); What does ``significantly more predictive'' mean? Do you mean ``example''? I dare say it doesn't. The foregoing makes no sense. You can only do a chi-squared (likelihood ratio) test in respect of **n; /archives/html/s-news/2004-08/msg00009.html (8,061 bytes)
11. Re: cluster simulation (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 30 Aug 2004 10:11:51 -0300 (ADT); (...) (...); /archives/html/s-news/2004-08/msg00176.html (7,310 bytes)
12. Re: Hidden Markov Model and Underflow problem (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 12 Jul 2004 14:13:48 -0300 (ADT); I have written a package which will fit hidden Markov models in which the observations have a discrete distribution specified nonparametrically. It could (I'm pretty sure) be ``fairly easily'' (famou; /archives/html/s-news/2004-07/msg00054.html (8,398 bytes)
13. Make problem in a library chapter. (score: 1): Author: rian Ripley <ripley@stats.ox.ac.uk>; Date: Wed, 23 Jun 2004 16:51:19 -0300 (ADT); I am trying to rebuild, (under Splus Version 6.1.2 Release 2 for Sun SPARC, SunOS 5.6 : 2002 ) a library chapter that I had running under Splus 3.? (back in the Middle Ages) and haven't touched withi; /archives/html/s-news/2004-06/msg00146.html (7,631 bytes)
14. Re: Make problem in a library chapter. (score: 1): Author: .uk>; Date: Wed, 23 Jun 2004 17:25:31 -0300 (ADT); Thanks! I edited the LIBRARY_FLAGS file in $(SHOME)/cmd and changed cc to gcc -- and bingo, it ran. I'm still a bit puzzled, but. Why does the damned thing need a C compiler at all? There is no c cod; /archives/html/s-news/2004-06/msg00147.html (7,363 bytes)
15. Re: Make problem in a library chapter. (score: 1): Author: .ca>; Date: Wed, 23 Jun 2004 17:41:44 -0300 (ADT); Thanks Bill. I already (at Don Kobayashi's suggestion) made a change of cc to gcc. Didn't do anything with the flags, but it seemed to work anyway. I'll go through the procedure more carefully now, a; /archives/html/s-news/2004-06/msg00148.html (6,902 bytes)
16. Different behaviour of unique(), R vs. Splus. (score: 1): Author: l Schmertmann <schmertmann@fsu.edu>; Date: Tue, 29 Jun 2004 12:20:52 -0300 (ADT); Apologies for the cross-posting, but I thought this snippet of info might be vaguely interesting to both lists. I did a **brief** search to see if this issue had previously been discussed and found n; /archives/html/s-news/2004-06/msg00167.html (8,086 bytes)
17. Re: Compare two regression models - one with a nonlinear constraint (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 2 Feb 2004 14:51:09 -0400 (AST); Can you give us some assurance that this is not a homework problem? cheers, Rolf Turner rolf@math.unb.ca; /archives/html/s-news/2004-02/msg00006.html (7,140 bytes)
18. Re: Compare two regression models etc. (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 2 Feb 2004 17:42:53 -0400 (AST); OK -- if it's thesis work, consider this to be a consultation. (No charge, of course!) It seems to me that the problem is quite simple; you need to consider two cases: Case 1: Z.bar not equal to 0. (; /archives/html/s-news/2004-02/msg00008.html (9,267 bytes)
19. Tenure track statistics position at UNB. (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Wed, 26 Jun 2002 15:05:13 -0300 (ADT); The Department of Mathematics and Statistics at the University of New Brunswick, Fredericton, N. B., Canada, is advertizing a tenure track position in Statistics, starting 1 July 2003. Please see htt; /archives/html/s-news/2002-06/msg00225.html (6,718 bytes)
20. Extreme value distributions (Long.) (score: 1): Author: Rolf Turner <rolf@math.unb.ca>; Date: Mon, 25 Mar 2002 15:41:01 -0400 (AST); This may not actually be an R/Splus problem, but it started off that way ..... ==+==+==+==+==+==+==+==+==+==+==+==+==+==+==+==+==+==+== Executive summary: == Simulations involving extreme value distr; /archives/html/s-news/2002-03/msg00185.html (25,814 bytes)

Current List: 1 - 20
Page: [1] [2] [3] [4] [5]