- 1. Re: aggregate, agg.sum,agg.length (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Thu, 12 Dec 2002 08:08:05 -0600
- Try the summarize function in the Hmisc library. It works much faster than aggregate for my needs. Kevin Wright --Original Message-- From: Steve Su [mailto:s.su@qut.edu.au] Sent: Thursday, December 1
- /archives/html/s-news/2002-12/msg00095.html (15,211 bytes)
- 2. Making a balanced data frame for prediction (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Fri, 13 Sep 2002 13:30:01 -0500
- I am fitting various models (aov, lme) on data frames that may not be balanced (with respect to the terms in the model). I would like to use predict(model.object, newdata) where newdata is a balanced
- /archives/html/s-news/2002-09/msg00097.html (8,814 bytes)
- 3. Re: strange quirk in Splus 6/Win2000 (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Wed, 3 Jul 2002 10:58:55 -0500
- As explained by this link, http://www.oreilly.com/catalog/jpower/bugfixes/ Windows 95/98/NT reserves filenames con.* and nul.* for it's own use. I had a similar problem when I tried to install a PERL
- /archives/html/s-news/2002-07/msg00015.html (10,901 bytes)
- 4. lme/gls/proc mixed (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Tue, 25 Jun 2002 19:25:40 -0500
- I am helping someone convert code from SAS proc mixed to S-Plus/R. After an hour of struggles, I am seeking some help. The proc mixed code is: proc mixed data = meps; class cross; model y = cross g1(
- /archives/html/s-news/2002-06/msg00217.html (7,296 bytes)
- 5. Superpose qqmath plots? (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Sat, 1 Dec 2001 10:58:49 -0600
- Is it possible to (reasonably) superpose qqmath plots? I would like to do something like: r<-data.frame(Grp=rep(c("A","B"),c(10,10)),y=c(rnorm(10),2*rnorm(10))) qqmath(~y,data=r, groups=Grp) The erro
- /archives/html/s-news/2001-12/msg00001.html (6,705 bytes)
- 6. Identical axis scales in parallel coordinate plot (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Wed, 28 Nov 2001 16:39:55 -0600
- I searched the S-news archive for a way to make the axes of a parallel coordinate plot have identical scales. I found no posted solution, so (having hacked a solution myself) post a solution for futu
- /archives/html/s-news/2001-11/msg00211.html (7,157 bytes)
- 7. Insightful's new web page (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Mon, 22 Oct 2001 15:25:03 -0500
- I visited the web page of Insightful today and was quite taken aback. After some wandering through the pages, my first reaction was, "Good grief, what hideously bad design!" Well, I'm probably experi
- /archives/html/s-news/2001-10/msg00138.html (7,242 bytes)
- 8. Slow trellis graphics in S-Plus 6 (Windows 95) (score: 1)
- Author: "Wright, Kevin" <kevin.wright@pioneer.com>
- Date: Thu, 27 Sep 2001 13:45:49 -0500
- i
- /archives/html/s-news/2001-09/msg00239.html (7,354 bytes)
- 9. Peculiar problem with varcomp (score: 1)
- Author: herneau@mayo.edu>
- Date: Fri, 17 Aug 2001 17:19:03 -0500
- I am using S-Plus 2000 (and S-Plus 6) on Windows 95. I have encountered what appears to be a strange scaling problem with the function varcomp. Broadly, I have these questions: 1. What are the relat
- /archives/html/s-news/2001-08/msg00139.html (11,573 bytes)
- 10. Calling lme (How DOES S-Plus work anyway) (score: 1)
- Author: "Wright, Kevin" <Kevin.Wright@pioneer.com>
- Date: Wed, 11 Jul 2001 23:19:37 -0500
- My system: S-Plus 2000 R 3 on Windows 95. I'm being plagued by a difference between aov and lme. I'm trying to manually build a model, then pass it into a function which calls lme. Things are not wo
- /archives/html/s-news/2001-07/msg00085.html (8,113 bytes)
- 11. Google searches of S-news archive (score: 1)
- Author: "Wright, Kevin" <Kevin.Wright@pioneer.com>
- Date: Fri, 29 Jun 2001 11:17:07 -0500
- When the S-news archive obtained a new search engine, I found it not to my liking. After some experimentation, I found that I could use Google for searching the archives. The process is fairly simple
- /archives/html/s-news/2001-06/msg00278.html (7,045 bytes)
This search system is powered by
Namazu