Mail Archive Search: <+from:pierre.joyet@basler.ch>

Results:

References: [ +from:pierre.joyet@basler.ch: 10 ]

Total 10 documents matching your query.

1. Degrees of freedom in smoothing splines (score: 1): Author: pierre.joyet@basler.ch; Date: Thu, 1 Feb 2001 09:26:46 +0100; Some days ago I asked a question about degrees of freedom of a spline in GAM. I thank Prof. Ripley for his answer, which didn't solve the problem (the question was rather unclear!). I have now better; /archives/html/s-news/2001-02/msg00002.html (7,952 bytes)
2. Degrees of freedom in GAM with smoothing spline (score: 1): Author: pierre.joyet@basler.ch; Date: Tue, 30 Jan 2001 10:08:05 +0100; I don't understand the following behaviour of S-Plus Version 2000, Release 1 (the data frame dat5 is defined at the end of this message); it is as if S-Plus always assumed df >= 5 in s(). weights = N; /archives/html/s-news/2001-01/msg00208.html (10,168 bytes)
3. [S] Generalized GAM (score: 1): Author: pierre.joyet@basler.ch; Date: Fri, 18 Feb 2000 16:05:56 +0100; Is there (in S-Plus 4.5) a direct way of fitting a model such as Z = s1(X) + Y * s2(X) where X and Y are continuous variables, * is the usual multiplication, and s1 and s2 are smooth functions to be; /archives/html/s-news/2000-02/msg00163.html (7,456 bytes)
4. [S] PRNG again... (score: 1): Author: pierre.joyet@basler.ch; Date: Wed, 19 Jan 2000 13:49:40 +0100; If I don't make a mistake, for Y Poisson(10)-distributed we have E[Y^4] = 16710. I tried "checking" that by simulation with S-Plus 4.5 Professional Release 1 on a Pentium II-PC, but mean(rpois(n, 10); /archives/html/s-news/2000-01/msg00136.html (7,601 bytes)
5. [S] References on geostatistics (score: 1): Author: Pierre Joyet <pierre.joyet@basler.ch>; Date: Mon, 02 Aug 1999 10:46:14 +0200; This question is not directly related to S-Plus, but S-News is the only way I have of contacting people in the right area. What would be a recommendable book (at the introductory level) about geostat; /archives/html/s-news/1999-08/msg00002.html (6,901 bytes)
6. [S] "..." in group methods (score: 1): Author: pierre.joyet@basler.ch (joyet pierre); Date: Fri, 18 Sep 1998 10:59:03 +0200; I have a class "dreieck" which inherits from class "data.frame". An element of this class is simply a data frame with 3 columns which represent a ragged array; the two first columns are the indices,; /archives/html/s-news/1998-09/msg00121.html (8,150 bytes)
7. [S] Numerical convolution (score: 1): Author: pierre.joyet@basler.ch (joyet pierre); Date: Thu, 30 Jul 1998 10:31:49 +0200; Using S-Plus 4.5, what is the most efficient numerical method for computing (in 1000 points, say) the density or probability function of X + Y, where X and Y are random variables whose joint density; /archives/html/s-news/1998-07/msg00211.html (7,046 bytes)
8. [S] SUMMARY: Numerical convolution (score: 1): Author: pierre.joyet@basler.ch (joyet pierre); Date: Fri, 31 Jul 1998 12:46:43 +0200; Yesterday I asked the following question: Many thanks to Andreas Krause, Bob Henery, Andrey Feuerverger and Peter Perkins for their helpful suggestions which I reproduce below. The fast Fourier trans; /archives/html/s-news/1998-07/msg00215.html (11,227 bytes)
9. [S] pch = Black point (score: 1): Author: pierre.joyet@basler.ch (joyet pierre); Date: Mon, 20 Apr 1998 16:29:17 +0200; In S-Plus 4 (release 3), for classical graphics in a graphsheet, is there a simple way of obtaining as plotting symbols moderately large black points like those that appear for example in many graphs; /archives/html/s-news/1998-04/msg00155.html (7,356 bytes)
10. [S] ordered (score: 1): Author: pierre.joyet@basler.ch (joyet pierre); Date: Thu, 30 Apr 1998 11:45:09 +0200; With S-Plus 4.0r3, when I type I get [1] 3 3 5 3 2 1 Levels (first 5 out of 10): [1] "0" "1" "2" "3" "4" 0 < 1 < 2 < 3 < 4 < 5 < 6 < 7 < 8 < 9 That's nice; but when I type I get Error in cut.default(; /archives/html/s-news/1998-04/msg00279.html (8,079 bytes)

Current List: 1 - 10
Page: [1]