similar to: [despammed] RE: multidimensional function fitting

You can use R objects, such as the return from gam, and the predict.gam function, from C. See the R extensions manual. Reid Huntsinger -----Original Message----- From: RenE J.V. Bertin [mailto:rjvbertin at despammed.com] Sent: Thursday, February 27, 2003 3:42 PM To: Wiener, Matthew Cc: r-help at stat.math.ethz.ch Subject: Re: [despammed] RE: [R] multidimensional function fitting On Thu, 27

Take a look at package mgcv. Hope this helps. --Matt -----Original Message----- From: RenE J.V. Bertin [mailto:rjvbertin at despammed.com] Sent: Thursday, February 27, 2003 1:39 PM To: r-help at stat.math.ethz.ch Subject: [R] multidimensional function fitting Hello, I have been looking around for how to perform a multidimensional, arbitrary function fit (in any case non-linear; more below),

>From: Patrick Connolly <p.connolly@hortresearch.co.nz> >To: "RenE J.V. Bertin" <rjvbertin@hotmail.com> >Subject: Re: [R] changing (core) function argument defaults? >Date: Thu, 18 Nov 2004 11:43:10 +1300 > >On Wed, 20-Oct-2004 at 07:48PM +0200, RenE J.V. Bertin wrote: > >|> Hello,

>From: Patrick Connolly <p.connolly@hortresearch.co.nz> >To: "RenE J.V. Bertin" <rjvbertin@hotmail.com> >Subject: Re: [R] changing (core) function argument defaults? >Date: Thu, 18 Nov 2004 11:43:10 +1300 > >On Wed, 20-Oct-2004 at 07:48PM +0200, RenE J.V. Bertin wrote: > >|> Hello,

You could use cut. The key calculation would be: w <- .05; eps <- 1e-5 breakpoints <- seq(min(kk), max(kk), .05) breakpoints <- floor( (breakpoints + (w/2) + eps) / w) * w values <- cut(kk, c(breakpoints, Inf), right = FALSE) values <- ordered(values) If you don't like the labels produced add lab = breakpoints as a cut arg. On 11/10/05, RenE J.V. Bertin

> Might not be it...but the '30 second lockup' triggered a memory... > > Could this be the 'looking for scheduled tasks' problem? > Here's a blurb from the > windows-net-mag site > (http://www.win2000mag.net/Articles/Index.cfm?ArticleID=24546) . Like Windows 2000, XP suffers from a little-documented browse delay. When you browse to a Windows 9x computer

Concerns: R 1.5.0 gui version, Windows (downloaded binary) and Linux (installed from sources). # Load the data from the attached file: kk<-read.table("__filename__", header=1) # attach the data: attach(kk) Snr<-factor(Snr) # fool around with a call to anova.glm(): anova.glm( aov( nFD~Type+size+Modality+Error(Snr/(Type+size+Modality)) ) ) # Error: object nFD not found # Well, I

On 6/23/05, RenE J.V. Bertin <rjvbertin at gmail.com> wrote: > Hello, > > I was just having a look at the aov function source code, and see that when the model used does not have an Error term, Helmert contrasts are imposed: > > if (is.null(indError)) { > ... > } > else { > opcons <- options("contrasts") >

Well... this may be a red herring after all, but it's an interesting one! It turns out I have got DYLD_LIBRARY_PATH set. This gets exported to R (or rather, affects the loader that loads the R process?), and somehow mixes up the way symbols are exported. Something that applies to launching R from the command line (/usr/bin/R) AND via LaunchServices (open -a R). Once I noticed that invoking

Rene, Look at ?format. Sean On Sep 17, 2004, at 9:21 AM, RenE J.V. Bertin wrote: > Hello, > > I can't seem to find the way to modify the precision with which > paste() prints its floating point numbers, more precisely the number > of decimal digits printed. This is apparently not controlled by > options( digits= ), and there is no appropriate argument to paste() >

Hi. I did write those functions, and sent them (I thought) to one of the R maintainers to see whether they would be appropriate for inclusion (because I'd seen some requests on the mailing lists). However, I'm happy to post them -- I should have thought of it before. WARNING: I've tested these functions on some data arising in my work and also on the USArrests data that comes with

Hi there, I'm trying to do a regression experiment on a multidimensional dataset where both x and y in the model are multidimensional vectors. I'm using R version 2.9.2, updated packages, on a Linux box. I've tried gausspr(), ksvm() and rvm(), and the models are computed fine, but I'm always getting the same error message when I try to use predict(): "Error in

Hello, Your way is much better than to mess with the dim attribute, like I did. But, "If you can create a data.frame or matrix that has the indices" Actually, it must be a matrix, indices can't be of type list. A way to avoid loops/apply altogether, and much faster, is the one creating K3 (K is the result from the op.) n <- 20 t2 <- system.time({ K2 <-

I am moving this to r-devel. The problem and solution below posted on r-help could have been a bit slicker if %*% worked with multidimensional arrays multiplying them so that if the first arg is a multidimensional array it is mulitplied along the last dimension (and first dimension for the second arg). Then one could have written: Tbar <- tarray %*% t(wt) / rep(wti, each = 9) which is a bit

Hi, anyone knows about any functions in R can get multidimensional integration not over a multidimensional rectangle (not adapt). For example, I tried the following function f(x,n)=x^n/n! phi.fun<-function(x,n) { if (n==1) { x }else{ integrate(phi.fun, lower=0, upper=x, n=n-1)$value } } I could get f(4,2)=4^2/2!=8, but failed in f(4,3)=4^3/3! Thanks Best, Lynette

Hello! the function apply() allows to compute the row sum of a multidimensional table but, when I try to merge the new column with the original table whith cbind(), the grouping varibles disapear and the new table is bidimensional. Knows anybody how to add row totals to a multidimensional table? Thanks in advance Juan Pablo _________________________________________________________________

On Friday 17 September 2004 13:52, RenE J.V. Bertin wrote: > Hello again, > > I am doing regressions (using panel.lmline() (and panel.abline( > rlm(...))) ) inside a panel method which I pass to bwplot(). > > What I would like to do is create a boxplot of categorised data > (binned on the independent variable), and superpose a regression line > which is calculated using the

A student of mine wants to use R to do some nonmetric multidimensional scaling. According to the R FAQ, there's a package called pcurve that computes multidimensional scaling solutions, but I was not able to locate it the contrib page (I am a Windows user with R version 1.4.1). Can anyone tell me whether it is possible to do nonmetric multidimensional scaling with R, and if so, how? John

Hello list. Another 'puzzle' for which I don't have a clean solution. Say I have a multidimensional object, e.g.: Mm<-matrix(1:6, nrow=2, dimnames=list(c("a","b"), c("g","h","i"))) And on the other hand I have a list Ind<-list("b","g") This holds, for each dimension, an indexer for that dimension. Now I would

This may be of interest: http://www.ijs.si/software/snprintf/ It looks like this version may well be more complete and conform the standards than the version that comes with R. BTW: I am currently patching unix/sys_std.c and modules/gtkconsole.c such that they don't store lines in the history that are identical to the previous line. Is there any interest in posting those patches here? RenE