Mail Archive Search: <+from:tjwade@uclink4.berkeley.edu>

Results:

References: [ +from:tjwade@uclink4.berkeley.edu: 15 ]

Total 15 documents matching your query.

1. problem with ts.plot (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Mon, 04 Dec 2000 17:06:41 -0800; I'm using version 4.5 on Windows system-- When I use "ts.plot" and I happen to have lines out of range, I get an error message and the program crashes. I need to exit and re-load S+ to continue, losi; /archives/html/s-news/2000-12/msg00033.html (6,919 bytes)
2. Re: Managing commands, scripts, workspaces (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Thu, 02 Nov 2000 06:40:07 -0800; 1) Is there an easy way to copy from the commands window into a script, ie without the prompts? I use the "History" window to do this, edit it and save it as a script. 2) Is there an option to set so; /archives/html/s-news/2000-11/msg00007.html (8,541 bytes)
3. Re: (pas d'objet) (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Thu, 30 Nov 2000 12:09:43 -0800; Try this: a<-c("NA","1", "NA", "5", "NA", "6","8","10") x<-c(1:length(a)) z<-z[a=="NA"] z should contain the positions of a where a is "NA" At 06:37 PM 11/30/2000 +0100, Marion VERDOIT wrote: Hello,; /archives/html/s-news/2000-11/msg00257.html (7,880 bytes)
4. Re: (pas d'objet) (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Thu, 30 Nov 2000 12:51:06 -0800; Try this: a<-c("NA","1", "NA", "5", "NA", "6","8","10") x<-c(1:length(a)) **z<-x[a=="NA"]** NOT z<-z[a=="NA"] At 06:37 PM 11/30/2000 +0100, Marion VERDOIT wrote: Hello, I have a vector a<-c("NA","1"; /archives/html/s-news/2000-11/msg00258.html (7,928 bytes)
5. [S] exact confidence interval for proportion (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Wed, 03 May 2000 11:43:37 -0700; Does anyone know if is there function in S+ to calculate exact confidence intervals for proportions based on binomial probability distribution (e.g. k successes in n trials). Are the confidence inte; /archives/html/s-news/2000-05/msg00023.html (7,146 bytes)
6. [S] exact ci for a proportion: summary (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Wed, 03 May 2000 16:47:13 -0700; Thanks very much to everyone who replied --I had several excellent answers (summarized below) all of which seem to get the job done, with some variation in the input options (some allow you to specif; /archives/html/s-news/2000-05/msg00025.html (17,105 bytes)
7. [S] lsfit or glm: specifying an intercept (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Sat, 05 Feb 2000 12:47:13 -0800; sorry if this question has been asked. I've only just subscribed. I would like to fit a simple least squares regression using lsfit or glm, but I want to force the least squares regression line and; /archives/html/s-news/2000-02/msg00040.html (7,671 bytes)
8. [S] graph display (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Mon, 14 Jun 1999 08:31:49 -0700; I am using S+ 4.5 for Windows. I have an unimportant but annoying problem with my graph display: for some reason when I create line graphs using different types of lines, they do not appear on my sc; /archives/html/s-news/1999-06/msg00117.html (7,871 bytes)
9. Re: [S] imputation (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Thu, 01 Apr 1999 10:40:13 -0800; The way we have commonly deal with non-detects is to use one-half of the limit (i.e. one half of 9=4.5) this is somewhat better than using the limit if you can assume that the non-detect values are r; /archives/html/s-news/1999-04/msg00003.html (9,481 bytes)
10. Re: [S] sign mail to s-news, please (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Wed, 07 Apr 1999 10:21:31 -0700; I have another request/question for this mailing list: Why is it that answers and/or replies are infrequently posted? Only when the asker summarizes and reposts the answers to we get a view of the re; /archives/html/s-news/1999-04/msg00038.html (8,899 bytes)
11. RE: [S] sign mail to s-news, please (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Wed, 07 Apr 1999 13:58:00 -0700; I didn't mean to create this flood of turmoil and discontent. I really think this mailing list is an incredible resource but I personally am always in need of programming help, tips and pointers and; /archives/html/s-news/1999-04/msg00049.html (9,008 bytes)
12. [S] attach data frame (score: 1): Author: "Timothy J. Wade" <tjwade@uclink4.berkeley.edu>; Date: Tue, 13 Apr 1999 17:29:35 -0700; I'm using S plus 4.5 for windows. I'm a little confused about the operation of the attach function. As I understand using attach(data.frame) should allow you to refer to the components of the data f; /archives/html/s-news/1999-04/msg00120.html (7,204 bytes)
13. [S] multiple graphs on a page (score: 1): Author: .edu>; Date: Thu, 11 Mar 1999 11:55:59 -0800; Hello S users, I have the following frustrating problem. I'm using Splus 4.5 for windows. I want to put multiple graphs on a page using the Windows functions and cut and paste. I proceed EXACTLY as d; /archives/html/s-news/1999-03/msg00130.html (7,144 bytes)
14. Re: [S] change values in a data frame conditionally (score: 1): Author: m.nl>; Date: Mon, 15 Mar 1999 07:19:58 -0800; I would do it like this: datframename[, "x2"]_ifelse(dataframename[, "x1"]==1 & dataframename[, "id"]==1, 1, dataframename[, "x2"]) essentially, this says if the dataframe columns "x1" and "id" equal; /archives/html/s-news/1999-03/msg00150.html (8,668 bytes)
15. Re: [S] survival analysis (score: 1): Author: c.nz>; Date: Tue, 16 Mar 1999 17:23:27 -0800; My only comment is that logistic regression would not be appropriate unless all observation times are equal. The fact that the Cox model explicitly incorporates different follow up times makes it bet; /archives/html/s-news/1999-03/msg00173.html (9,811 bytes)

Current List: 1 - 15
Page: [1]