And for the sake of completeness :) I'll also mention the S-PLUS 6 for Windows will be able to do the same for C++ programs (when it's released). A draft white paper discussing the new CONNECT/C++ f
Sv4 strings aren't officially supported in S-PLUS 6.0 for Unix or Windows. String objects aren't discussed in the S-PLUS documentation, and the on-line help page for string() states "Function string
This has been fixed in S-PLUS 6. This is the new behaviour on both Windows and Unix: x a y X1 1 a 1 NA NA NA NA NA1 NA NA NA NA2 NA NA NA The all-NA rows are superfluous, but I believe this is the in
Bert is absolutely correct: win.test[,3] is a vector, and so the column name is lost. But on the other hand, win.test[,3,drop=F] is a data frame with a single column named "three", so win.new <- cbi
I can't explain what's going on except that it looks like a bug to me. In any case, it appears to be fixed in S-PLUS 6. Here's what I get for both the released Unix versions and the pre-beta Windows
Longtime s-news subscribers may recall that support for S-PLUS for DEC Alpha machines had been discontinued following S-PLUS 5. Well, the good news for Compaq Alpha (as it's now known) users is that
Inspired by Don's post, I've made a similar graphic available on the S-PLUS pages at: http://www.s-plus.com/graphlets/votes.html I've used S-PLUS 6's new Graphlets feature to create the interactive g
I just tried this (entering ;; on the command line) on a fresh install of S-PLUS 2000 R3 and it seems that this problem was fixed. (Although the fix might have been more coincidence than design, as t
S-PLUS 2000 Release 3 is now available from the S-PLUS website at http://www.splus.com/resources/download/ This release was built primarily to address Windows 2000 compatibility issues. A complete li
A large number of users have downloaded and installed S-PLUS 2000 R3 without problems, so I would suspect that this problem is specific to this particular install. David, I will pass your message on
[I'd been holding off posting this until I could update the FAQ page, and it seems David got in first :).] Thanks to David providing us with a clear example of how to reproduce his problem with the R
One of the major changes in S-PLUS 6 for Windows is the introduction of the S version 4 (Sv4) language engine. It includes features such as memory mapping, reference counting, and block updating of f
It's time to register for the Statistical Modeling and Computation in Finance Conference being held September 28th at the Marriott Marquis in New York City. We are excited by the response from the S-
MathSoft S-PLUS invites s-news readers to attend the one-day Statistical Modeling and Computation in Finance conference being held September 28th in New York City, where you can - Attend presentation
this question, but I thought I would ask it anyway. Suppose I run a glm (poisson regression) using the statement: glm.obj <- glm(formula = Y ~ offset(log(
RAM I have 20 years of hourly data. I need to compute the 0-24 hour autocorrelation of one parameter. I first need to get the data in some sort of time se
members to the S-PLUS User's conference in Seattle, October 18-20. Anyone who was at the conference in New Orleans last year will know that this is a great
le solution to this problem. I have three columns in a data frame which I want to use to create a 2 dimensional scatter plot. Here are the details: 1) Xvar
lighted to have people offer improvements to my code, that's the way to learn things. In this particular instance, I suspect that for long lists with large
lf Turner, Shawn Bowles, and Bill Venables for their help. Apparently, this was not the challenge I thought it was! It came down to two algorithms: good <-