Search String: Display: Description: Sort:

Results:

References: [ +from:SmithSJ@mar.dfo-mpo.gc.ca: 35 ]

Total 35 documents matching your query.

1. Re: How to do a plot with two Y-axis? Thanks! (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 09 Nov 2004 14:22:55 -0400
Take a look at pages 371--372 of the Programmer's Guide (Vers. 6.x). Stephen Smith --Original Message-- From: Shuxia Yu [mailto:ysxzh@163.com] Sent: November 9, 2004 2:18 PM To: s-news@lists.biostat.
/archives/html/s-news/2004-11/msg00062.html (9,364 bytes)

2. Re: clustered boxplots (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Fri, 14 May 2004 08:41:33 -0300
There are two ways you get the boxplot you want. For your example: y1<-rnorm(10,0,1) y2<-rnorm(10,1,1) y3<-rnorm(10,3,1) Simply type: Or define: y.data<-c(y1,y2,y3) group.data<-c(rep(1,10),rep(2,10),
/archives/html/s-news/2004-05/msg00075.html (7,896 bytes)

3. Variance functions (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Fri, 05 Mar 2004 11:54:11 -0400
Windows XP, SPLUS 6.2 (Rev. date 12 Jan Build 6713), nlme version 3.3.2. I am trying to understand how to use variance functions in nlme() I had raised this issue on the list last September but was u
/archives/html/s-news/2004-03/msg00070.html (13,785 bytes)

4. FW: Variance functions (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 09 Mar 2004 12:55:27 -0400
Further to what I have written below, varFixed() seems to work as expected for lme(). Stephen Smith --Original Message-- From: Smith, Stephen [mailto:SmithSJ@mar.dfo-mpo.gc.ca] Sent: March 5, 2004 11
/archives/html/s-news/2004-03/msg00094.html (14,564 bytes)

5. Variance Functions (again) (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 09 Mar 2004 20:10:02 -0400
Apparently, the original posting was rejected. Further to what I have written below, varFixed() seems to work as expected for lme(). Stephen Smith --Original Message-- From: Smith, Stephen Sent: Marc
/archives/html/s-news/2004-03/msg00101.html (10,893 bytes)

6. Re: hurdle or two-step models (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Fri, 26 Mar 2004 15:06:55 -0400
You could also check out the following papers: @ARTICLE{Stefansson96, author = "Gunnar Stef\'{a}nsson", title = "Analysis of groundfish survey abundance data: combining the GLM and delta approaches",
/archives/html/s-news/2004-03/msg00222.html (9,129 bytes)

7. NLME technical question (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Wed, 03 Sep 2003 10:14:24 -0300
I am trying to fit a nlme model for growth curves with weights=varfixed and the grouping variable being set to locations that the animals were sampled at. I get the following error: + t0 ~ 1, random
/archives/html/s-news/2003-09/msg00013.html (13,510 bytes)

8. Bug in browser? (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Wed, 02 Jul 2003 16:20:18 -0300
I am trying to debug the following program using browser() in SPLUS 6.1 using Windows XP. prepare.data<-function(data){ sample.unique<-unique(data$sample.no) out<-vector(mode="list",length=0) for(i i
/archives/html/s-news/2003-07/msg00012.html (14,024 bytes)

9. Re: Extracting the Hat matrix from a least-squares fit (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 02 Jul 2002 09:09:38 -0300
With respect to using the correct back-transformation when assuming a lognormal regression model, you may be interested in: @ARTICLE{BraduMundlak70, author = "D. Bradu and Y. Mundlak", title = "{Esti
/archives/html/s-news/2002-07/msg00007.html (11,570 bytes)

10. Re: Summary and Exhortation: Greek in Axis Labels (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 22 May 2001 20:27:00 -0300
While I don't have a daily need for including Greek on axis labels and have used the psfrag route once or twice, having this ability within S-PLUS traditional graphics (i.e., command line) would be
/archives/html/s-news/2001-05/msg00223.html (11,242 bytes)

11. Re: Summary and Exhortation: Greek in Axis Labels (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Wed, 23 May 2001 15:02:46 -0300
This approach will be problematic when working with trellis plots. Trellis plots created using line commands do not fare very well when changed into objects -- borders and strip labels no longer line
/archives/html/s-news/2001-05/msg00243.html (11,460 bytes)

12. difference between summary minimum and min() --- a bug? (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Fri, 06 Oct 2000 16:57:31 -0300
It is late on a Friday afternoon before Thanksgiving weekend and I am trying to run some tests on the network version of SPLUS for our Informatics group. To test some calculations I ran summary() on
/archives/html/s-news/2000-10/msg00052.html (9,571 bytes)

13. RE: difference between summary minimum and min() --- a bug? (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 10 Oct 2000 08:44:28 -0300
Prof Ripley: I must have missed something in my education but I had only expected rounding in the decimal places. By the way, in Canada we were always taught dd/mm/yyyy. Stephen Smith
/archives/html/s-news/2000-10/msg00077.html (14,492 bytes)

14. has predict.gam changed? (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Wed, 25 Oct 2000 09:33:32 -0300
I am having a strange problem with predict.gam. I have fit a linear model to some data (some details below) where the independent variables are factors. I wish to make safe predictions from the resul
/archives/html/s-news/2000-10/msg00207.html (9,113 bytes)

15. [S] offsets (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 25 Apr 2000 09:50:45 -0300
Can anyone direct me to some material on using offsets in generalized linear models? I thought I understood what they were supposed to do but after playing around with a Poisson model with (linearly
/archives/html/s-news/2000-04/msg00221.html (7,636 bytes)

16. FW: [S] offsets (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 25 Apr 2000 10:56:18 -0300
Someone just brought in a copy of the third edition of V&R and after looking through it I gave myself a quick slap to the head. I had included the offset as an argument in glm() not as part of the fo
/archives/html/s-news/2000-04/msg00224.html (8,738 bytes)

17. FW: [S] offsets, more (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Tue, 25 Apr 2000 14:01:46 -0300
Okay, I am back to square one here. After including my offset in the formula I still don't see a difference between the analysis before adding the offset and after adding the offset. Any insights her
/archives/html/s-news/2000-04/msg00227.html (8,589 bytes)

18. [S] hockey sticks (score: 1)
Author: "Smith, Stephen" <SmithSJ@mar.dfo-mpo.gc.ca>
Date: Thu, 27 Apr 2000 09:32:55 -0300
With respect to who used the "hockey stick" appellation first -- I found the following references in Current Index to Statistics. 78PrInStMa 25 29- 40 J Threshold values as safe levels and the hockey
/archives/html/s-news/2000-04/msg00243.html (7,883 bytes)

19. RE: [S] RTF and S-Plus integration with other products (score: 1)
Author: SmithSJ@mar.dfo-mpo.gc.ca
Date: Mon, 22 Nov 1999 13:17:20 -0400
[Smith, Stephen] I would like to dip my oar into this contentious sea as well. I work in a Government Department in which computing (Informatics as it is officially referred to) is more about control
/archives/html/s-news/1999-11/msg00229.html (8,994 bytes)

20. RE: [S] sample size!!! (score: 1)
Author: SmithSJ@mar.dfo-mpo.gc.ca
Date: Wed, 18 Aug 1999 13:04:42 -0300
Yours is not an SPLUS question. Determine sample size for what criterion? That is, minimum variance or minimum cost or a function of both. Check with the standard sampling texts, e.g., @BOOK{Cochran7
/archives/html/s-news/1999-08/msg00163.html (9,572 bytes)

Current List: 1 - 20
Page: [1] [2]
This search system is powered by Namazu