Mail Archive Search: <+from:york@orca.akctr.noaa.gov>

Results:

References: [ +from:york@orca.akctr.noaa.gov: 12 ]

Total 12 documents matching your query.

1. [S] Behavior of predict in Splus 2000 (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Mon, 4 Oct 1999 15:05:08 -0700 (PDT); I'm finding that the predict function behaves very diffently in Splus 2000 than in previous versions of Splus. Here is a simple example in which I first define a data frame and fit a linear model. Th; /archives/html/s-news/1999-10/msg00030.html (9,289 bytes)
2. Re: [S] Behavior of predict in Splus 2000 (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Tue, 5 Oct 1999 07:10:07 -0700 (PDT); Problem solved. Thanks to Bill Dunlap and Chuck Taylor at Statsci who suggested I had a masked version of model.matrix.default in one of my directories. Removing it fixed the behavior of predict. Ann; /archives/html/s-news/1999-10/msg00038.html (10,310 bytes)
3. [S] predict function for nls (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Wed, 13 Jan 1999 14:49:54 -0800 (PST); I can't seem to find an Splus function for doing prediction for a non-linear least squares model -- in Splus 3.3 for Windows and splus 3.4 for unix? Are there any to be shared? Thanks, Anne ~~~~~~~~~; /archives/html/s-news/1999-01/msg00065.html (7,234 bytes)
4. [S] Postscript printing in Windows (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Fri, 13 Nov 1998 12:14:29 -0800 (PST); There have been several posts concerning the quality of postscript output from Splus using the MS drivers. There is an Splus postscript function that still exists in Splus 3.3 and that function seems; /archives/html/s-news/1998-11/msg00082.html (8,539 bytes)
5. [S] Inverse of true.file.name() ?? (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Fri, 2 Oct 1998 14:49:49 -0700 (PDT); "true.file.name" is a function in the Windows version of Splus that provides the dos filename of an splus object. e.g. [1] "__115" "dum" I would like to have a function that gives the splus object na; /archives/html/s-news/1998-10/msg00017.html (7,791 bytes)
6. Re: [S] dll.load and Digital Visual Fortran (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Mon, 27 Jul 1998 07:44:46 -0700 (PDT); * *Does anyone know when will be released the S+ version for Linux? * * * Alberto G. Murta Many of us are waiting. If more software like S+ were available on Linux, it would contributed toward giving; /archives/html/s-news/1998-07/msg00196.html (9,070 bytes)
7. Re: [S] white backgrounds under trellis (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Fri, 15 May 1998 07:32:22 -0700 (PDT); transparant is color = -999999 in both unix and dos. I have found this especially useful in superimposing a legend on a graph when you don't want to blot out the graph underneath (background = -99999; /archives/html/s-news/1998-05/msg00123.html (8,940 bytes)
8. Re: [S] white backgrounds under trellis (correction) (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Fri, 15 May 1998 13:58:24 -0700 (PDT); Earlier today I sent a message saying that one could use color = -999999 to get transparent output on a graph. This is not correct. What does work under the old legend command to get a transparent ba; /archives/html/s-news/1998-05/msg00128.html (7,425 bytes)
9. Re: [S] apropos (was re:unlist) (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Thu, 21 May 1998 07:06:41 -0700 (PDT); l; /archives/html/s-news/1998-05/msg00200.html (10,979 bytes)
10. Re: [S] RE: Documentation (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Thu, 28 May 1998 08:25:32 -0700 (PDT); l; /archives/html/s-news/1998-05/msg00303.html (8,979 bytes)
11. Re: [S] strange p-value for Fisher exact's test (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Tue, 24 Mar 1998 14:48:04 -0800 (PST); Not a Pentium problem. These were run in Splus 3.4 on an SGI with an Irix operating system: Fisher's exact test data: matrix(c(1, 1, 70, 70), nc = 2) p-value = 1 alternative hypothesis: two.sided Fis; /archives/html/s-news/1998-03/msg00254.html (11,048 bytes)
12. [S] Beta-binomial family (score: 1): Author: Anne York <york@orca.akctr.noaa.gov>; Date: Thu, 19 Feb 1998 08:53:30 -0800 (PST); Does anyone have Splus functions to estimate parameters with glm for a beta-binomial distribution using glm? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Anne E. York National Marine Mammal Laboratory Seattle; /archives/html/s-news/1998-02/msg00175.html (7,652 bytes)

Current List: 1 - 12
Page: [1]