Search String: Display: Description: Sort:

Results:

References: [ +from:tjagger@blarg.net: 19 ]

Total 19 documents matching your query.

1. Re: use hist to plot frequency distribution (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Tue, 18 Oct 2005 00:38:17 -0600
You need to span the range of your data. That is if rx<-range(x) ## and your set of breaks is mybreaks then you get the error if rx[1] < min(mybreaks) || rx[2] > max(mybreaks) returns T. So you migh
/archives/html/s-news/2005-10/msg00056.html (8,772 bytes)

2. Re: Comparing matrix and vector (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Mon, 11 Jul 2005 11:01:19 -0600
Tristan, I think you mean to find all rows of X whose 4th column contains a value from test, try using %in%, Y<-X[X[,4]%in%test,] FYI: get("%in%") function(x, table) !is.na(match(x, table, nomatch =
/archives/html/s-news/2005-07/msg00050.html (8,578 bytes)

3. Re: Concerns about SPLUS version 7 (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Tue, 12 Jul 2005 12:12:05 -0600
I use a university wide license for version 6.2 and 7.0, but even in the academic environment license management is difficult. While I could use the features of large data objects, I don't want to de
/archives/html/s-news/2005-07/msg00079.html (14,178 bytes)

4. Re: Batch update of datetime stamp. (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Mon, 18 Jul 2005 15:50:01 -0600
Yes by name is needed, try the following: clean.all <-function(id) { obj.name<-paste("station",id,sep="") value<-clean.dates(get(where=1,obj.name)$weather.date) eval(substitute(X$weather.date<-VALUE
/archives/html/s-news/2005-07/msg00106.html (10,831 bytes)

5. Re: This must be a simple task... (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Sat, 23 Jul 2005 10:19:36 -0600
Hi, maybe I am missing something here. Columns in a data frame all have the same length, so what are you replacing the say three values of 1.5 with? If you mean that you have a list and order is not
/archives/html/s-news/2005-07/msg00123.html (10,760 bytes)

6. Re: Applying vector functions to dataframes (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Tue, 26 Jul 2005 10:47:35 -0600
Good morning. I have written many functions that perform similarly. If you can use a single function on the whole matrix, using matrix mathematics then that is the fastest. For example the unbiased c
/archives/html/s-news/2005-07/msg00138.html (9,769 bytes)

7. Re: i'm perplexed (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Wed, 11 May 2005 19:08:16 -0600
,
/archives/html/s-news/2005-05/msg00061.html (15,683 bytes)

8. Re: Interpreting glmmpql results. (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Thu, 12 May 2005 13:12:54 -0600
r
/archives/html/s-news/2005-05/msg00077.html (16,180 bytes)

9. Re: Padding out a vector of characters to a constant width (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Mon, 16 May 2005 19:50:22 -0600
m
/archives/html/s-news/2005-05/msg00098.html (10,390 bytes)

10. Re: Question in Time Series (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Wed, 25 May 2005 12:54:50 -0600
c
/archives/html/s-news/2005-05/msg00155.html (13,822 bytes)

11. Re: Question in Time Series, fitting correlated data. (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Thu, 26 May 2005 00:23:02 -0600
i
/archives/html/s-news/2005-05/msg00159.html (23,328 bytes)

12. Re: tricky element pulling question (score: 1)
Author: .com>
Date: Thu, 7 Apr 2005 12:52:19 -0600
: s-news-owner@li
/archives/html/s-news/2005-04/msg00044.html (17,837 bytes)

13. Re: subscripting (score: 1)
Author: .edu>
Date: Thu, 7 Apr 2005 13:46:22 -0600
ector/list/matrix
/archives/html/s-news/2005-04/msg00046.html (9,093 bytes)

14. Re: permutation (score: 1)
Author: .com>
Date: Thu, 14 Apr 2005 16:01:50 -0600
.5 1.0 1.5 2.0 3.
/archives/html/s-news/2005-04/msg00102.html (19,251 bytes)

15. Using aggregate (score: 1)
Author: o.au>
Date: Sun, 24 Apr 2005 21:44:44 -0600
riginal Message--
/archives/html/s-news/2005-04/msg00149.html (10,601 bytes)

16. Re: pi.fit option in predict.glm (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Fri, 11 Mar 2005 10:39:33 -0700
Hello, in the function predict.glm, while it seems the calculation of link and response confidence limits are correct those for the prediction intervals are not since the variance component due to th
/archives/html/s-news/2005-03/msg00053.html (12,273 bytes)

17. Re: Derivatives of incomplete gamma function (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Sun, 27 Mar 2005 01:29:19 -0700
Sonia, I don?t have any fast methods, but here is a start. There are two ways to do it. The first way is just to take a small incremental say delta h, and try simple derivatives. For example: dIG.da<
/archives/html/s-news/2005-03/msg00143.html (11,637 bytes)

18. Re: bootstrapping of multiple datasets (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Wed, 28 Jun 2006 17:37:13 -0600
You can use the missing data library. If you have the imputations, you can do the following from the missing library: 1. Run you linear model for each set of imputations and put into a list: fit<-lis
/archives/html/s-news/2006-06/msg00082.html (10,644 bytes)

19. Re: yaaq: yet another apply question (score: 1)
Author: "Thomas Jagger" <tjagger@blarg.net>
Date: Fri, 28 Jul 2006 12:44:21 -0600
Use eval(parse(paste(text=...))) form. lapply(xxx, function(x) try(eval(parse(text=paste("library(help=",x,")") )))) sections without help files. This generates a lot of readme.txt notepad windows (
/archives/html/s-news/2006-07/msg00077.html (8,410 bytes)


This search system is powered by Namazu