[R] distance in the function kmeans
My thread broke as I write this at home and there were no new messages on this subject after I got home. I hope this still reaches interested parties. There are several methods that find centroids (means) from distance data. Centroid clustering methods do so, and so does classic scaling a.k.a. metric multidimensional scaling a.k.a. principal co-ordinates analysis (in R function cmdscale the means are found in C function dblcen.c in R sources). Strictly this centroid finding only works with Euclidean distances, but these methods willingly handle any other dissimilarities (or distances). Sometimes this results in anomalies like upper levels being below lower levels in cluster diagrams or in negative eigenvalues in cmdscale. In principle, kmeans could do the same if she only wanted. Is it correct to use non-Euclidean dissimilarities when Euclidean distances were assumed? In my field (ecology) we know that Euclidean distances are often poor, and some other dissimilarities have better properties, and I think it is OK to break the rules (or `violate the assumptions'). Now we don't know what kind of dissimilarities were used in the original post (I think I never saw this specified), so we don't know if they can be euclidized directly using ideas of Petzold or Simpson. They might be semimetric or other sinful dissimilarities, too. These would be bad in the sense Uwe Ligges wrote: you wouldn't get centres of Voronoi polygons in original space, not even non-overlapping polygons. Still they might work better than the original space (who wants to be in the original space when there are better spaces floating around?) The following trick handles the problem euclidizing space implied by any dissimilarity meaasure (metric or semimetric). Here mdata is your original (rectangular) data matrix, and dis is any dissimilarity data: tmp <- cmdscale(dis, k=min(dim(mdata))-1, eig=TRUE) eucspace <- tmp$points[, tmp$eig > 0.01] The condition removes axes with negative or almost-zero eigenvalues that you will get with semimetric dissimilarities. Then just call kmeans with eucspace as argument. If your dis is Euclidean, this is only a rotation and kmeans of eucspace and mdata should be equal. For other types of dis (even for semimetric dissimilarity) this maps your dissimilarities onto Euclidean space which in effect is the same as performing kmeans with your original dissimilarity. Cheers, jari oksanen -- Jari Oksanen, Oulu, Finland __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Re: Problem: creating shared objects using lapack and blas
To solve my own problem, it seems better to follow instructions in "Writing R Extensions" to build a library. I am making progress. Thanks, Zhu Wang Statistical Science Department Southern Methodist University Dallas, TX 75275-0332 On Fri, 2004-05-28 at 13:11, Zhu Wang wrote: > I am having trouble with creating shared objects with Fortran files, > which use numerical libraries Lapack and Blas. I have read the section > in "Writing R Extensions" but could not find what I needed and I am not > at the stage to create a library. > > What I did was > > $R CMD SHLIB --output=car file1.f file2.f ... file50.f -llapack -lblas > > No error message so far, > > But > > $>dyn.load("./car") > Error in dyn.load(x, as.logical(local), as.logical(now)) : > unable to load shared library "/home/zwang/R/pkg/car/src/./car": > /home/zwang/R/pkg/car/src/./car: undefined symbol: slamch_ > > I think the way I used to link Lapack and Blas was not correct, even > though it worked fine when I used for running Fortran code before. > > Thanks for any advice. > > Zhu Wang > > Statistical Science Department > Southern Methodist University > Dallas, TX 75275-0332 -- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] GLMM error in ..1?
Is this the new experimental lme4 (version 0.6-x) ? If so, this is due to an error in our use of method dispatch. It has been fixed in the development version, and there should be a new release in a few days. On Friday 28 May 2004 19:32, Spencer Graves wrote: > I'm trying to use GLMM in library(lme4), R 1.9.0pat, updated > just now. I get an error message I can't decipher: > > library(lme4) > set.seed(1) > n <- 10 > N <- 1000 > DF <- data.frame(yield=rbinom(n, N, .99)/N, nest=1:n) > fit <- GLMM(yield~1, random=~1|nest, family=binomial, data=DF, > weights=rep(N, n)) > > Error in eval(expr, envir, enclos) : ..1 used in an incorrect > context, no ... to look in > > Is there something I can do to get past this short of working > through the code for GLMM line by line? > > Also, I gather GLMM does not currently accept > > binomial(link="cloglog"): > > fit <- GLMM(1-yield~1, random=~1|nest, > > family=binomial(link="cloglog"), > > + data=DF,weights=rep(N, n)) > Error in getClass(thisClass) : "family" is not a defined class > Error in GLMM(1 - yield ~ 1, random = ~1 | nest, family = > binomial(link = "cloglog"), : > S language method selection got an error when called from > internal dispatch for function "GLMM" __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] glmmPQL:
I'm getting a strange error from glmmPQL. Consider the following sample code: set.seed(8) N. <- 1000 z <- rnorm(N.) pr.good <- exp(-1e-4*exp(2+2*z)) quantile(pr.good) DF. <- data.frame(yield=rbinom(N., N., pr.good)/N., Offset=rep(-10, N.), nest=1:N.) fit <- glmmPQL(fixed=1-yield~offset(Offset), random=~1|nest, family=binomial(link="cloglog"), data=DF., weights=rep(N., N.)) Using R 1.9.0pat, Windows 2000, with MASS updated just a few hours ago, I get the following error: iteration 1 iteration 2 Error in logLik.reStruct(object, conLin) : NA/NaN/Inf in foreign function call (arg 3) Any suggestions? Best Wishes, Spencer Graves __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to pass defined function with more than one arguments to apply?
Not sure what you intend with regard to length but to get a logical vector indicating which rows equal a particular vector: f1 <- function(tb, row) apply(tb,1,function(x)all(x==row)) # or without using apply: f2 <- function(tb, row) colSums( abs( (t(tb) - row) ) ) == 0 # or in terms of a general compare function: f3 <- function(tb, row, compare = function(x)all(x==row)) apply(tb,1,compare) row <- c(0,3,1) f1(tb,row) f2(tb,row) f3(tb,row) # If you want the number of matching rows: length(which(f1(tb,row))) etc. anoly hotmail.com> writes: : : Dear all: : I meet a problem of apply function. I have a matrix called tb : >tb :V1 V2 V3 : 1 0 3 1 : 2 1 4 0 : 3 0 3 0 : 4 0 4 0 : 5 0 3 1 : 6 1 4 1 : 7 1 1 0 : 8 1 3 0 : 9 0 1 1 : 10 0 3 1 : : I hope to get the number of row that match c(0,3,1) : I do this way: : >length(apply(t(tb) = = c (0,3,1), 2, all)) : I defined a funtion, compare<-function(vector1, vector2){...}. For example, : compare(1:3, 1:3) will return TRUE. compare(1:3,2:4) return FALSE. : Then I hope to call apply(tb,1,compare). But this can not work, because : apply only pass one argument to compare function. Does anyone know how to : solve this problem? : : Thanks so much. : Anoly : : __ : R-help stat.math.ethz.ch mailing list : https://www.stat.math.ethz.ch/mailman/listinfo/r-help : PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html : : __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] GLMM error in ..1?
I'm trying to use GLMM in library(lme4), R 1.9.0pat, updated just now. I get an error message I can't decipher: library(lme4) set.seed(1) n <- 10 N <- 1000 DF <- data.frame(yield=rbinom(n, N, .99)/N, nest=1:n) fit <- GLMM(yield~1, random=~1|nest, family=binomial, data=DF, weights=rep(N, n)) Error in eval(expr, envir, enclos) : ..1 used in an incorrect context, no ... to look in Is there something I can do to get past this short of working through the code for GLMM line by line? Also, I gather GLMM does not currently accept binomial(link="cloglog"): > fit <- GLMM(1-yield~1, random=~1|nest, family=binomial(link="cloglog"), + data=DF,weights=rep(N, n)) Error in getClass(thisClass) : "family" is not a defined class Error in GLMM(1 - yield ~ 1, random = ~1 | nest, family = binomial(link = "cloglog"), : S language method selection got an error when called from internal dispatch for function "GLMM" > Thanks, spencer graves __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] "privileged slots"
On 28 May 2004 at 15:09, Wolski wrote: > The simplest way to find the document is > ?checkSlotAssignmen Thanks, I finally found it. > A function which is described there and which Matthias Kohl mentioned > in his e-mail You will find it also using the html search form of the > html documentation searching for RClassUtils. Yes, I found that one as well. Now I see what your original concern was. I didn't realize that you were talking about the "check" argument of "slot<-". Anyway, it's still true that "@<-" adds some overhead simply because it's a wrapper around "slot<-". > Its true that the document is a little hidden what I havent noticed > (Sorry Martin).The functions described in it are not listed in the > 00index.html file of the methods package. (Why!?)I found the document > by the first trial by typing ?checkSlotAssignment. I think most of them should be considered "internal" and may be subject to change. Also, the text says: "Most of them are rarely suitable to be called directly." > There was a lot of discussion going on atabout S4 at the useR > confernce. Much more than at this list.Maybee because the only > understandable text about S4 about wich i knew is "Programming with > Data" J. Chambers. I haven't found anything in "Programming with Data" about performance issues or differences concerning "slot<-" and "@<-". Indeed, I can't find any hints that both should be different by any means (except that one can specify anything that evaluates to a slot name when using "slot<-"). At least, the language specification doesn't suggest such a difference in my opinion. My copy of S (S-PLUS 6.1 for Windows) shows that both are implemented identically in S: > get("@<-") # or get("slot<-") function(object, name, value) .Call("do_AT_GETS", object, name, value) So the difference between the two is R specific. > (The descriptions to the method package is for me > quite hard to read.) Maybee also because no one knews how big exactly > is the performance sacrifice caused by the additional > features.Therefore, this "less" official document is ofsome value > because it allows to make some benchmarks. Well, the "check" argument is documented in ?slot. Reading this, one can guess what it does without having to read the docs for checkSlotAssignment(). I agree that the description of checkSlotAssignment() is quite interesting because of this: "For privileged slots (those that can only be set by accesor functions defined along with the class itself), ..." Unfortunately, I haven't found any hints that privileged slots are implemented; and I wouldn't use them anyway as long as they aren't "officially" documented. > I was running some test cases to figure out how many(much?) time one > can gain by setting check=FALSE check=TRUE. > ... > It means that a type check takes about 0.0002 seconds on a P4 3000 > Windows XP with hyperthreding. Its not much. It means thats no reason > to be scared of S4 because of performance.The overhad is even less > 0.2 Thanks for doing some tests here. The results don't surprise me; I didn't expect the operations performed by checkSlotAssignment() to be extraordinarily expensive. > Last but not least: There where also some discussions about having > theOPTIONto pass arguments to function by reference. That might be useful in some situations. I wonder how well this fits into the original language concept ? BFN, Torsten [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] vector normal to a plane
Thomas Lumley u.washington.edu> writes: : : W is looking for the vector cross product (a specifically : three-dimensional object important in physics and engineering). The : crossproduct() function provides a matrix product so that crossprod(x,y) : is t(x)%*%y, something completely different. : : For the three-dimensional case you could define the cross and dot products : : "%x%"<-function(a,b) {c(a[2]*b[3]-a[3]*b[2], -a[1]*b[3]+a[3]*b[1], : a[1]*b[2]-a[2]*b[1])} Alternately, you could get the cofactors from solve: cross3 <- function(a,b) { m <- cbind(a,b,1) solve(m)[3,]*det(m) } __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Converting data frame to array?
Dear Thomas, In fact, the more common way to fit a linear regression in R is to use variables in a data frame (or list) along with a model formula specifying the model. All of this is explained in the Introduction to R manual that is distributed with R: see, in particular, Sec. 6.3 on data frames, Sec. 7 on reading data from files, and Sec. 11 on statistical models. Given two data frames, say d1 and d2, the first containing, e.g., observations on variables x1 and x2 and the second on y, one could do lm(y ~ x1 + x2, data=c(x1, x2)) or lm(y ~ x1 + x2, data=data.frame(x1, x2)). That said, it's not altogether clear to me what it is that you're trying to do. Are there 10 observations on 300 variables in the first data frame, constituting the predictors, and 10 observations on 1 variable in the second data frame, constituting the response? If so, you have many more predictors than observations, and it's not reasonable to perform a regression. Of course, I may not have this straight. I hope this helps, John > -Original Message- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of TAPO > (Thomas Agersten Poulsen) > Sent: Friday, May 28, 2004 2:11 PM > To: [EMAIL PROTECTED] > Subject: [R] Converting data frame to array? > > Dear List, > > Please bear with a poor newbee, who might be doing > everything backwards (I was brought up in pure math). > > I want to make a simple multi-linear regression on a > set of data. I did some expreiments, and if X is a 4 by 2 > array and Y is a 4 by > 1 array, I can do a linear regression by lm(y~x). > > Now I have a tab-delimited text file with 10 rows of > 300 measurements and an other file with 10 rows of one value > each. When I read in those files using read.delim(), I get > data frames, and apparently I can no longer do the > multi-linear regression. > > Is there a way to convert the data frames into arrays, > or am I going the wrong way about this? > > Sincerely > Thomas Poulsen > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] vector normal to a plane
W is looking for the vector cross product (a specifically three-dimensional object important in physics and engineering). The crossproduct() function provides a matrix product so that crossprod(x,y) is t(x)%*%y, something completely different. For the three-dimensional case you could define the cross and dot products "%x%"<-function(a,b) {c(a[2]*b[3]-a[3]*b[2], -a[1]*b[3]+a[3]*b[1], a[1]*b[2]-a[2]*b[1])} "%.%%<-function(a,b) sum(a*b) It would make sense, of course, to check that the arguments actually were vectors of length 3. -thomas __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] vector normal to a plane
Uwe Ligges wrote: [EMAIL PROTECTED] wrote: Hi All, (I have a degree in math, but I am too embarassed to ask my colleagues, so here goes:) I would like to get a vector normal (orthogonal) to a plane formed by two other vectors. In matlab I do this: v1 = [.4, .6, .8]; v2 = [.9, .7, .2]; nn = cross(v1,v2) (gives ~[-.48, .65, -.24] Huh? I don't have access to Matlab. Can you tell me how cross() is defined in Matlab (it's not obvious to me - at least not at 10:40 pm - how can anything get negative)? My apologies - and thanks to Rolf Turner who told me the truth (well, I anticipated that it's getting too late this friday evening). OK, I don't think what you are looking for is defined as a function in R. Anyway, for the 3D case it's pretty easy to write the result in one line of code ... Uwe if I do R> cross(v1, v2), I get .94. Huh? Ha! ;-) crossprod(v1, v2): 0.4*0.9 + 0.6*0.7 + 0.8*0.2 = 0.94 Uwe Ligges Thanks for all your help, again. W __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Converting data frame to array?
TAPO (Thomas Agersten Poulsen) wrote: Is there a way to convert the data frames into arrays, or am I going the wrong way about this? This is possible, but the behaviour depends on the datatype, e.g. numeric or character. Simply look for ?as.matrix or ?as.array Thomas P. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] vector normal to a plane
[EMAIL PROTECTED] wrote: Hi All, (I have a degree in math, but I am too embarassed to ask my colleagues, so here goes:) I would like to get a vector normal (orthogonal) to a plane formed by two other vectors. In matlab I do this: v1 = [.4, .6, .8]; v2 = [.9, .7, .2]; nn = cross(v1,v2) (gives ~[-.48, .65, -.24] Huh? I don't have access to Matlab. Can you tell me how cross() is defined in Matlab (it's not obvious to me - at least not at 10:40 pm - how can anything get negative)? if I do R> cross(v1, v2), I get .94. Huh? Ha! ;-) crossprod(v1, v2): 0.4*0.9 + 0.6*0.7 + 0.8*0.2 = 0.94 Uwe Ligges Thanks for all your help, again. W __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] vector normal to a plane
Hi All, (I have a degree in math, but I am too embarassed to ask my colleagues, so here goes:) I would like to get a vector normal (orthogonal) to a plane formed by two other vectors. In matlab I do this: v1 = [.4, .6, .8]; v2 = [.9, .7, .2]; nn = cross(v1,v2) (gives ~[-.48, .65, -.24] if I do R> cross(v1, v2), I get .94. Huh? Thanks for all your help, again. W __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] (Correction) vector normal to a plane
if I do R> cross(v1, v2), I get .94. Huh? Meant to say "R> crossprod(v1, v2)" Sorry __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] vector normal to a plane
Hi! In which library is this function? I cant find it using help.search() on my R installation. Eryk *** REPLY SEPARATOR *** On 5/28/2004 at 12:22 PM [EMAIL PROTECTED] wrote: >Hi All, > >(I have a degree in math, but I am too embarassed to ask my colleagues, >so here goes:) > >I would like to get a vector normal (orthogonal) to a plane formed by >two other vectors. In matlab I do this: > >v1 = [.4, .6, .8]; v2 = [.9, .7, .2]; nn = cross(v1,v2) (gives ~[-.48, >.65, -.24] > >if I do R> cross(v1, v2), I get .94. Huh? > >Thanks for all your help, again. > >W > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] About creating an import library
Hi all, First of all, I found R.exp. Thanks. The following is a paragraph I copied from Venables and Ripleys book S programming on Page242. If the code calls entry points in the R engine, create an import library from the file R.exp in the source-package distribution by lib /def: R.exp /machine: ix386 /out: Rdll.lib and add Rdll.lib to the link command or add it to the libraries to be used in the IDE. My problems are: 1. Why do we need an import library? (a very basic question) 2. What is R.exp? 3. After I copied the R.exp to C:\ and run lib /def: R.exp /machine: ix386 /out: Rdll.lib, the error message showed Cannot open Rdll.lib 4. I am not clear about add Rdll.lib to the link command or add it to the libraries to be used in the IDE. Thanks Rui Wang Phone: (403)220-4501 Email: [EMAIL PROTECTED] Department of Mathematics and Statistics University of Calgary [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Converting data frame to array?
Dear List, Please bear with a poor newbee, who might be doing everything backwards (I was brought up in pure math). I want to make a simple multi-linear regression on a set of data. I did some expreiments, and if X is a 4 by 2 array and Y is a 4 by 1 array, I can do a linear regression by lm(y~x). Now I have a tab-delimited text file with 10 rows of 300 measurements and an other file with 10 rows of one value each. When I read in those files using read.delim(), I get data frames, and apparently I can no longer do the multi-linear regression. Is there a way to convert the data frames into arrays, or am I going the wrong way about this? Sincerely Thomas Poulsen __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] dotchart questions
Instead of dotchart, you might try a lattice dotplot after restructuring the data. FACT1 <- vector("character", 15) for(i in 1:15){ FACT1[i] <- paste(rep(LETTERS[i], 20), collapse ="")} mydata <- data.frame(FACT1 = rep(FACT1, 9), FACT2 = rep(c("E", "F", "G"), c(15, 15, 15)), RESPVAL = runif(45*3), RESPVAR = rep(c("V1", "V2", "V3"), c(45, 45, 45))) library(lattice) trellis.device(width=11, height=8.5, new = FALSE, col = FALSE, bg = "white") dotplot(FACT1 ~ RESPVAL | RESPVAR * FACT2, data = mydata, between = list(x = 0.5, y = 0)) hope this helps, Chuck Cleland Richard Valliant wrote: I am trying to put 3 dotcharts side-by-side with minimal space between each. Each chart is for a different variable, but the vertical axes are the same. I want to have vertical axis labels on the lefthand chart but no vertical axis labels on the other two. Plus, I would like very little space between charts 1 & 2 and between charts 2 & 3. I have one approach but am not too happy with the results. A stripped down version of the code is below. Matrix A has the row labels (which are about 20 characters long in the real data). Matrix A1 has no row labels. A <- matrix(1:12, ncol=3, byrow=F) A1 <- A dimnames(A) <- list(letters[1:4], letters[5:7]) dimnames(A1) <- list(rep("",4), rep("",3)) par(mfrow = c(1,3), mgp = c(1.75, 1, 0.5), mar = c(3, 0, 1.5, 0), oma = c(2,2,2,2)) dotchart(A) dotchart(A1) dotchart(A1) Two questions: (1) Is there a way with par or something else to make the charts closer together. (2) With a lot of rows (45 in the real data) is there a way to control row spacing (other than getting rid of some rows). -- Chuck Cleland, Ph.D. NDRI, Inc. 71 West 23rd Street, 8th floor New York, NY 10010 tel: (212) 845-4495 (Tu, Th) tel: (732) 452-1424 (M, W, F) fax: (917) 438-0894 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] intro statistics course in San Francisco bay area
I'm in San Francisco looking for a basic statistics course... using R. I'll be using R for real work later on, so I'd prefer not to have to do the coursework with some other software. All leads are appreciated... Thanks, - JD __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Problem: creating shared objects using lapack and blas
I am having trouble with creating shared objects with Fortran files, which use numerical libraries Lapack and Blas. I have read the section in "Writing R Extensions" but could not find what I needed and I am not at the stage to create a library. What I did was $R CMD SHLIB --output=car file1.f file2.f ... file50.f -llapack -lblas No error message so far, But $>dyn.load("./car") Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library "/home/zwang/R/pkg/car/src/./car": /home/zwang/R/pkg/car/src/./car: undefined symbol: slamch_ I think the way I used to link Lapack and Blas was not correct, even though it worked fine when I used for running Fortran code before. Thanks for any advice. Zhu Wang Statistical Science Department Southern Methodist University Dallas, TX 75275-0332 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] gauss.hermite?
Thanks to Andy Liaw, Marc Schwartz, Dimitris Rizopoulos and Tamas Papp for quick replies. Before I sent the email, I found that it was written by Jim Lindsey, but I could not find further information on him, so I appreciate the links. Also, the Advanced Statistical Computing course notes look quite interesting. In particular, the function "gaulag" providing "Gauss-Laguerre" may help with a related problem, taking expected values with respect to gamma distributions. (Laguerre polynomials are orthogonal with respect to the a gamma distribution, and perform the same role as Hermite polynomials relative to a normal distribution.) Thanks again. spencer graves Tamas Papp wrote: On Fri, May 28, 2004 at 09:04:53AM +0200, Dimitris Rizopoulos wrote: Dear Spencer, In Dr. Gray's course notes for Advanced Statistical Computing (Appendix of Chapter 6) there are some functions for computing abscissas and corresponding weights for several Gaussian integration rules: http://icommons.harvard.edu/~hsph-bio248-01/Lecture_Notes/ Also have a look at @Book{judd98, author = {Judd, Kenneth L}, title ={Numerical methods in economics}, publisher ={MIT Press}, year = 1998 } especially one of the early chapters ("integration" is in its title, but I don't have the book at the moment), which covers theory and practice of numerical integration and quadrature rules. Implementing these in R has been on my TODO list for a while, so I would be happy to cooperate on this (but only in July or later). Best, Tamas __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How could I find R.exp?
On Fri, 28 May 2004 11:52:58 -0600, "Rui" <[EMAIL PROTECTED]> wrote : >Hi all, > >I try to create an import library for using dll. Following the >instruction in the readme.package, I used the command lib /def:R.exp >/out:Rdll.lib, however, the error message showed Cannot open file >R.exp. I downloaded the source code R-1.9.0.tgz, but I can not find >R.exp. Thanks! It's not part of the source code, it's produced when R is built, and should be in the src/gnuwin32 directory, as long as you chose to install "Source Package Installation Files". Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How could I find R.exp?
Hi all, I try to create an import library for using dll. Following the instruction in the readme.package, I used the command lib /def:R.exp /out:Rdll.lib, however, the error message showed Cannot open file R.exp. I downloaded the source code R-1.9.0.tgz, but I can not find R.exp. Thanks! Rui Phone: (403)220-4501 Email: [EMAIL PROTECTED] Department of Mathematics and Statistics University of Calgary [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] RE: SCO & R
The intellectual property questions raised by the GNU General Public License (GPL) are under active discussion in PDF Solutions, where Sundar Dorai-Raj and I work. Sundar and I have recommended linking PDF commercial software to R. However, we have to be careful how we do this, because the GPL requires that the source for "all derivative works" be made public. Unfortunately, it seems unclear exactly what constitutes a "derivative work", although the GPL states that it does not apply to "independent and separate works ... when you distribute them as separate works." Does anyone have information on this, beyond the groklaw and Forbes items already mentioned? For example, are there good examples to follow, or are all the published case studies negative like those already been mentioned? It seems to me that a commercial company could comply with the GPL without jeopardizing its ability to pay employees and stockholders by distributing its software in two, separately installed pieces: The first would provide the base, commercial, proprietary software under a commercial license. The second would include all components distributed under the GPL license, and would add capabilities to the first. I think something like this would benefit both the R Foundation and commercial software distributors. However, our legal department has yet to rule on this, and our management is not eager to move without the blessing of our corporate legal. Therefore, more solid information, published case studies, etc., would help. Thanks, spencer graves Berton Gunter wrote: My $.02 (sorry, couldn't resist). This is but one skirmish in the whole vexatious battle on "intellectual property rights," which extends from software to many other venues, one of the most interesting of which is patents on human genes. An example I heard on the radio this morning was eHarmony.com patenting their algorithm for matchmaking, which, the commentator said, is based on a 431 item questionnaire (yet another reason to celebrate my long marriage; 431 questions -- yikes!) ! Needless to say, the potential fodder for editorializing both for and against this intrusion of "scientific method" into romance seems limitless. Do you think support vector machines or neural nets lurk in the background of their "algorithm"? Anyway, the only predictions that I think it safe to make are: (1) We have only begun to fight -- and, yes, seeming samaritans (or is it samurai?) like R and GPL may yet have to engage; (2) advances in technology will only turn the heat up and pose yet more problems, as we move relentlessly from "thingy" inventions running our world to 'idea-y" inventions running it. Matchmaking, anyone? (Thanks for your indulgence). Cheers, Bert "Siddique, Amer" wrote: SCO is grasping for straws. and is now gasping. vexatious lawsuits will only drive so far before puttering out. eventually the curtain gets pulled back on glorified attempts at racketeering. cheers. Message: 76 Date: Thu, 27 May 2004 21:08:12 -0500 From: Deepayan Sarkar <[EMAIL PROTECTED]> Subject: Re: [R] SCO & R To: [EMAIL PROTECTED] Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="iso-8859-1" On Thursday 27 May 2004 20:18, Greg Tarpinian wrote: Actually, it was very recent. I pulled the electronic version of the article from the Forbes website: I would think twice before taking forbes articles too seriously. For instance, they seem to think that protecting intellectual property rights are all right when it's some big commercial company doing the protecting, but not when they are on the other side of the table. See this article from last year: http://www.forbes.com/2003/10/14/cz_dl_1014linksys.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Bert Gunter Non-Clinical Biostatistics Genentech MS: 240B Phone: 650-467-7374 "The business of the statistician is to catalyze the scientific learning process." -- George E.P. Box __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Sorting Data?
Am 28.05.2004 15:18:58 schrieb(en) Jonathan Baron: >Also the other things. > >I have to test the hypothesis: Does a daugther answer the questions >(semantic differential) more equal the own mother and more different to >the mothers of the other daugthers. I don't think this is a trivial problem at all, so I am hesitant to offer advice. I see now that you reall do want a matrix, where you have mothers in the columns and daughters in the rows, and distance (difference, similarity) measures in the cells. Perhaps you have several such measures, so you want a three-dimensional array. Hm. I don't know. I tried the following: Creating a new Variable by: Fbg[['F.gesamt']] <- rowSums(Fbg[11:27], na.rm=TRUE) Which contains everything from F01 to F17 as sum(). Than I do the thing to split mothers from daugthers as you told me. The next step i get: dist(Muetter['F.gesamt'] - Toechter['F.gesamt']) Which gives me some output, where I am not sure what it is. ;-( But it looks like the thing I want, but I think I don't really have the right data inside. You might do something like this. First define a function to measure your distance, like Itemdist <- function(x,y) {sum(abs(x[-(1:4)]-y[-(1:4)]))} Dists <- matrix(NA,nrow(Daughters),nrow(Mothers)) for (i in 1:nrow(Daughters) {for (j in 1:nrow(Mothers)) {Dists[i,j] <- Itemdist(Daughters[i,],Mothers[j,])}} Hm. I get some syntax errors inside this. If you want you can have my small sample I'm now createing a source file about. There I have 12 mother daugther pairs. I will get about 20 pairs within the next two months, but then I will have my source file (loading by source("file. R")) ready. The small sample is just to work now. Thanks for your help, Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] orca binary?
I cross compiled it. If someone wants to donate a pre-configured Windows computer to my research group, I could test it. Else, someone else will have to. I can't. It is http://www.analytics.washington.edu/~rossini/rorca/rorca.zip best, -tony [EMAIL PROTECTED] (A.J. Rossini) writes: > Uwe Ligges <[EMAIL PROTECTED]> writes: > >> Laura Holt wrote: >> >>> Dear R People: >>> Is there a binary version for rorca, please? >> >> >> Given I recall correctly, I tried to build rorca for Windows one or >> two years ago without success. But maybe Tony (In this context I >> recall that I'm still owing him a pizza) knows of a binary? > > Hah. I should swing by and collect next week. > > Seriously, you should be able to do a source install or a quick build > (I probably could do one later today using the Linux cross-compiler), > since it's just raw R code. > > If someone does it before I do, let me know. > > best, > -tony > > -- > [EMAIL PROTECTED]http://www.analytics.washington.edu/ > Biomedical and Health Informatics University of Washington > Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center > UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable > FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email > > CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}} > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > -- [EMAIL PROTECTED]http://www.analytics.washington.edu/ Biomedical and Health Informatics University of Washington Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}} __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Is there a way to represent the ... argument in signature?
Have a look at rbind.zoo in the zoo package for an example. Wolski molgen.mpg.de> writes: : : Hi! : : I guess that it cant work. but maybee I am wrong. : I would like to define a function rbind. : > rbind : function (..., deparse.level = 1) : { : which works only for objects of class Myclass : : Is it possible to use something like : setMethod("rbind",signature(...="Myclass",deparse.level="numeric") : This gives an error. : Or should I use : rbind.Myclass : : Eryk. : : Dipl. bio-chem. Eryk Witold Wolski MPI-Moleculare Genetic : Ihnestrasse 63-73 14195 Berlin 'v' : tel: 0049-30-83875219 / \ : mail: wolski molgen.mpg.de---W-W http://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Generate a sequence of random integer values
On 28-May-04 Ted Harding wrote: > > (1:100)[sort(runif(100),index.return=TRUE)] > > 37 99 22 66 12 36 90 48 55 45 98 59 92 20 46 23 2 74 > 80 5 63 27 56 60 51 76 39 87 19 96 29 15 18 100 52 3 > 70 8 67 42 43 57 91 21 25 78 41 1 34 68 77 9 72 73 > 11 89 83 94 65 13 14 81 24 58 35 44 10 79 50 49 6 28 > 17 33 4 71 82 31 75 30 85 93 26 69 38 84 32 54 64 53 > 88 61 86 97 7 62 95 47 16 40 > > but there may be a slicker way ... > Ted. Of course there is (if what you want in random order is (1:100)) since that's what sort(runif(100),index.return=TRUE)$ix gives you anyway, and the initial (1;100) is redundant! However, if it's some other set X of 100 items that you want in random order then X[sort(runif(100),index.return=TRUE)$ix] is indeed what you need. Ted. E-Mail: (Ted Harding) <[EMAIL PROTECTED]> Fax-to-email: +44 (0)870 167 1972 Date: 28-May-04 Time: 17:57:56 -- XFMail -- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Is there a way to represent the ... argument in signature?
Hi! I guess that it cant work. but maybee I am wrong. I would like to define a function rbind. > rbind function (..., deparse.level = 1) { which works only for objects of class Myclass Is it possible to use something like setMethod("rbind",signature(...="Myclass",deparse.level="numeric") This gives an error. Or should I use rbind.Myclass Eryk. Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Negative binomial glm and dispersion
Using R 1.8.1, and the negative binomial glm implemented in MASS, the default when using anova and a chi-square test is to divide the deviance by the estimated dispersion. Using my UNIX version of S-plus (v 3.4), and the same MASS functions, the deviances are *not* divided by the estimated dispersion. Firstly, I'm wondering if anyone can enlighten about the correct procedure (I thought the F-test was more appropriate when dispersion is estimated)? Secondly, after a bit of muddling with the negative binomial pdf, I concluded that, like for the Poisson, phi is actually 1. This result is borne out by simulations. Is this correct? # an example in R 1.81 with library(MASS) y<-rnegbin(n=100,mu=1,theta=1) x<-1:length(y) model<-glm(y~x,family=neg.bin(1)) summary(model)$dispersion [1] 1.288926 anova(model,test='Chisq") #... Df Deviance Resid. Df Resid. Dev P(>|Chi|) NULL99102.038 x 10.18598101.853 0.705 # But the "real" chi-square probability is 1-pchisq(0.185,1) [1] 0.667 Thanks in advance, Dan Daniel Kehler Dept. of Biology Dalhousie University, B3H 4J1 Halifax, Nova Scotia, Canada Office: LSC 800 email: [EMAIL PROTECTED] phone: 902 494 3910 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] dotchart questions
I am trying to put 3 dotcharts side-by-side with minimal space between each. Each chart is for a different variable, but the vertical axes are the same. I want to have vertical axis labels on the lefthand chart but no vertical axis labels on the other two. Plus, I would like very little space between charts 1 & 2 and between charts 2 & 3. I have one approach but am not too happy with the results. A stripped down version of the code is below. Matrix A has the row labels (which are about 20 characters long in the real data). Matrix A1 has no row labels. A <- matrix(1:12, ncol=3, byrow=F) A1 <- A dimnames(A) <- list(letters[1:4], letters[5:7]) dimnames(A1) <- list(rep("",4), rep("",3)) par(mfrow = c(1,3), mgp = c(1.75, 1, 0.5), mar = c(3, 0, 1.5, 0), oma = c(2,2,2,2)) dotchart(A) dotchart(A1) dotchart(A1) Two questions: (1) Is there a way with par or something else to make the charts closer together. (2) With a lot of rows (45 in the real data) is there a way to control row spacing (other than getting rid of some rows). I checked the email archives but found no answers there or in the help file. I have R.1.8.1 Windows XP, 512 MB RAM Thanks, Richard Valliant, Ph.D. University of Maryland Joint Program for Survey Methodology 1218 Lefrak Hall College Park MD 20742 (301)-405-0932 FAX: (301) 314-7912 [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Generate a sequence of random integer values
On 28-May-04 Brian Davenhall wrote: > I'm trying to generate a sequence of random integer values. I've > tried to combine the random (r) and the sequence (seq) functions > but this approach does not work. For example, if I use the > following command: > >> a <- seq(1:100) >> a > [1] 1 2 3 4 5 6 7 8 9 etc. > > This is a good start, but what I really want is something that > would look like this instead > > [1] 3 96 45 67 8 24 99 63 8, etc. > > where the integer numbers between 1 and 100 are randomly chosen. > > Any help would be great, I've found workarounds in other stat packages, > but would prefer to do this in R. (1:100)[sort(runif(100),index.return=TRUE)$ix] 37 99 22 66 12 36 90 48 55 45 98 59 92 20 46 23 2 74 80 5 63 27 56 60 51 76 39 87 19 96 29 15 18 100 52 3 70 8 67 42 43 57 91 21 25 78 41 1 34 68 77 9 72 73 11 89 83 94 65 13 14 81 24 58 35 44 10 79 50 49 6 28 17 33 4 71 82 31 75 30 85 93 26 69 38 84 32 54 64 53 88 61 86 97 7 62 95 47 16 40 but there may be a slicker way ... Ted. E-Mail: (Ted Harding) <[EMAIL PROTECTED]> Fax-to-email: +44 (0)870 167 1972 Date: 28-May-04 Time: 17:33:52 -- XFMail -- __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to insert a bitmap in a grph device
Dirk Eddelbuettel wrote: > IIRC the pixmap package on CRAN helps with that. Ah, I see, ?addlogo is much easier than grid in this case, as one sees in a (slightly modified) version of the help example: x <- read.pnm(system.file("pictures/logo.ppm", package="pixmap")[1]) plot(sample(1:100)) for (i in 1:7) addlogo(x, px=c(0, (101/77)*11), py=c((i-1)*11, i*11), asp=1) Thank you Thomas P. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to insert a bitmap in a grph device
Luis Rideau Cruz wrote: If I have a plot and I want to insert a bitmap in the same devicehow to do it? Yesterday we have learned, how to load and plot jpeg images (see: Is it possible to read jpeg files into R?) Now you need a way to insert this into another figure. For this purpose the grid and gridBase packages can be applied, which is explained in an article from Paul Murrel in R-News 02/2003 and in his lecture at the useR! conference, see: http://cran.r-project.org/doc/Rnews/ and http://www.ci.tuwien.ac.at/Conferences/useR-2004/Keynotes/ useR! Thomas P. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] How to insert a bitmap in a grph device
On Fri, May 28, 2004 at 04:58:58PM +0100, Luis Rideau Cruz wrote: > If I have a plot and I want to insert a bitmap in the same devicehow to do it? IIRC the pixmap package on CRAN helps with that. Dirk -- FEATURE: VW Beetle license plate in California __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] how to pass defined function with more than one arguments to apply?
Hi! ?apply ...: optional arguments to 'FUN'. you can pass more arguments to your function. length(apply(tb,comare,c(0,3,1))) Sincerely Eryk. *** REPLY SEPARATOR *** On 5/28/2004 at 12:05 PM anoly wrote: >Dear all: >I meet a problem of apply function. I have a matrix called tb >>tb > V1 V2 V3 >1 0 3 1 >2 1 4 0 >3 0 3 0 >4 0 4 0 >5 0 3 1 >6 1 4 1 >7 1 1 0 >8 1 3 0 >9 0 1 1 >10 0 3 1 > >I hope to get the number of row that match c(0,3,1) >I do this way: >>length(apply(t(tb) = = c (0,3,1), 2, all)) > I defined a funtion, compare<-function(vector1, vector2){...}. For >example, >compare(1:3, 1:3) will return TRUE. compare(1:3,2:4) return FALSE. >Then I hope to call apply(tb,1,compare). But this can not work, because >apply only pass one argument to compare function. Does anyone know how to >solve this problem? > >Thanks so much. >Anoly > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] how to pass defined function with more than one arguments to apply?
Dear all: I meet a problem of apply function. I have a matrix called tb >tb V1 V2 V3 1 0 3 1 2 1 4 0 3 0 3 0 4 0 4 0 5 0 3 1 6 1 4 1 7 1 1 0 8 1 3 0 9 0 1 1 10 0 3 1 I hope to get the number of row that match c(0,3,1) I do this way: >length(apply(t(tb) = = c (0,3,1), 2, all)) I defined a funtion, compare<-function(vector1, vector2){...}. For example, compare(1:3, 1:3) will return TRUE. compare(1:3,2:4) return FALSE. Then I hope to call apply(tb,1,compare). But this can not work, because apply only pass one argument to compare function. Does anyone know how to solve this problem? Thanks so much. Anoly __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] RE: SCO & R
My $.02 (sorry, couldn't resist). This is but one skirmish in the whole vexatious battle on "intellectual property rights," which extends from software to many other venues, one of the most interesting of which is patents on human genes. An example I heard on the radio this morning was eHarmony.com patenting their algorithm for matchmaking, which, the commentator said, is based on a 431 item questionnaire (yet another reason to celebrate my long marriage; 431 questions -- yikes!) ! Needless to say, the potential fodder for editorializing both for and against this intrusion of "scientific method" into romance seems limitless. Do you think support vector machines or neural nets lurk in the background of their "algorithm"? Anyway, the only predictions that I think it safe to make are: (1) We have only begun to fight -- and, yes, seeming samaritans (or is it samurai?) like R and GPL may yet have to engage; (2) advances in technology will only turn the heat up and pose yet more problems, as we move relentlessly from "thingy" inventions running our world to 'idea-y" inventions running it. Matchmaking, anyone? (Thanks for your indulgence). Cheers, Bert "Siddique, Amer" wrote: > SCO is grasping for straws. and is now gasping. vexatious lawsuits will only > drive so far before puttering out. eventually the curtain gets pulled back > on glorified attempts at racketeering. cheers. > > Message: 76 > Date: Thu, 27 May 2004 21:08:12 -0500 > From: Deepayan Sarkar <[EMAIL PROTECTED]> > Subject: Re: [R] SCO & R > To: [EMAIL PROTECTED] > Message-ID: <[EMAIL PROTECTED]> > Content-Type: text/plain; charset="iso-8859-1" > > On Thursday 27 May 2004 20:18, Greg Tarpinian wrote: > > > Actually, it was very recent. I pulled the electronic version of the > > article from the Forbes website: > > I would think twice before taking forbes articles too seriously. For > instance, they seem to think that protecting intellectual property > rights are all right when it's some big commercial company doing the > protecting, but not when they are on the other side of the table. See > this article from last year: > > http://www.forbes.com/2003/10/14/cz_dl_1014linksys.html > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Bert Gunter Non-Clinical Biostatistics Genentech MS: 240B Phone: 650-467-7374 "The business of the statistician is to catalyze the scientific learning process." -- George E.P. Box __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] How to insert a bitmap in a grph device
If I have a plot and I want to insert a bitmap in the same devicehow to do it? Thank you Luis Ridao Cruz Fiskirannsóknarstovan Nóatún 1 P.O. Box 3051 FR-110 Tórshavn Faroe Islands Phone: +298 353900 Phone(direct): +298 353912 Mobile: +298 580800 Fax: +298 353901 E-mail: [EMAIL PROTECTED] Web:www.frs.fo __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Build R-1.9.0 with static libraries ?
I am still unable to build a binary for R (linux Red Hat 9 distribution) which would be self-consistent (without shared objects and use of dynamics librairies). I would be very grateful if someone could explain and/or help me... so please set MAIN_LD as it suggests (and that's why I quoted the whole thing for you to read!) OK, I've set MAIN_LD to 'gcc -Bstatic' but now 'make' fails for the following reason: Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library "$BuildDir/library/methods/libs/methods.so": $BuildDir/library/methods/libs/methods.so: undefined symbol: R_GlobalEnv Execution halted Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library "$BuildDir/library/stats/libs/stats.so": $BuildDir/library/stats/libs/stats.so: undefined symbol: R_GlobalEnv Execution halted Error in dyn.load(x, as.logical(local), as.logical(now)) : unable to load shared library "$BuildDir/library/tools/libs/tools.so": $BuilDir/library/tools/libs/tools.so: undefined symbol: R_NaString Error: couldn't find function ".installPackageNamespaceInfo" In addition: Warning message: package tools in options("defaultPackages") was not found Execution halted So I'm afraid I'm still missing a flag somewhere... I'm sorry but I'm not used compiling such packages so I cannot figure out where I made a mistake. Thanks again for your patience :-( Melanie -- Me'lanie PELEGRINI-ISSAC tel : (33 0)1 53 82 84 20 Unite' 483 INSERM fax : (33 0)1 53 82 84 48 9, quai Saint-Bernard email :[EMAIL PROTECTED] Bat C 6e etage 75005 PARIS __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] RE: SCO & R
SCO is grasping for straws. and is now gasping. vexatious lawsuits will only drive so far before puttering out. eventually the curtain gets pulled back on glorified attempts at racketeering. cheers. Message: 76 Date: Thu, 27 May 2004 21:08:12 -0500 From: Deepayan Sarkar <[EMAIL PROTECTED]> Subject: Re: [R] SCO & R To: [EMAIL PROTECTED] Message-ID: <[EMAIL PROTECTED]> Content-Type: text/plain; charset="iso-8859-1" On Thursday 27 May 2004 20:18, Greg Tarpinian wrote: > Actually, it was very recent. I pulled the electronic version of the > article from the Forbes website: I would think twice before taking forbes articles too seriously. For instance, they seem to think that protecting intellectual property rights are all right when it's some big commercial company doing the protecting, but not when they are on the other side of the table. See this article from last year: http://www.forbes.com/2003/10/14/cz_dl_1014linksys.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] simulate dependent probabilities
I need to simulate from a random process and am not sure how to go about it. The process is the probability of an event occuring between a pair of points on a line. (This probability is between 0 and 0.5). I have estimates of these probabilities for a series of points, their standard errors and the correlation matrix (which is AR(1)). Eg (for 4 points) estimated prob (q): 0.1163 0.1280 0.0698 standard error: 0.0320 0.0288 0.0259 asymptotic correlation matrix: 1. -0.0880 1. 0. -0.0739 1. The vector q is used in a further analysis, treated as known. I would like to simulate alternative vectors q, which could be used in the further analysis in order to generate some empirical confidence interval. But I don't know where to start with such simulation. (In practice, q has about 50 elements). Although I know how to use cholesky decomposition to simulate dependent variables from a MVN distribution, I am stuck on two counts here: - the distribution for q - how to incorporate the dependence into the simulation. I would appreciate any suggestions. Chris. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] axis.POSIXct: Datetime data and plotting
At 2:10 PM -0400 5/27/04, Barrowman, Nick wrote: But is there any easy way to avoid this problem? It is confusing and inconvenient that axis.POSIXct(1,d,format=f) and axis.POSIXct(3,at=d,format=f) give seemingly contradictory scales. In one case you provide the locations of the tick marks, in the other case you let the function decide the locations. Why, in general, would anyone expect the two to be the same? (And that applies regardless of whether we're dealing with datetime values; the regular axis() function doesn't always give me tick marks at locations that I prefer.) Gabor Grothendieck's suggestion to use the Date class instead of POSIXt is a good one. -Don At 2:10 PM -0400 5/27/04, Barrowman, Nick wrote: I've run into a problem with the datetime axis generated by axis.POSIXct. It appears a similar issue was discussed in October 2003 under the subject line "datetime data and plotting" (see https://stat.ethz.ch/pipermail/r-help/2003-October/039071.html), but I wasn't able to determine whether there is a straightforward solution. The code below produces a graph with apparently contradictory date labels on the top and bottom axes: f <- "%Y/%m/%d" d <- as.POSIXct(strptime(c("2003/1/15","2003/1/16", "2003/1/17","2003/1/18","2003/1/19"),format=f)) plot(d,1:length(d),axes=F);box() axis(2) abline(v=d[3]) axis.POSIXct(1,d,format=f) axis.POSIXct(3,at=d,format=f) On my copy of R (I have appended the version information below my signature), this produces a graph with a vertical line that lines up with "2003/01/17" on the top axis, but not on the bottom axis. The issue seems to relate to time zones: here's what d looks like on my computer: d [1] "2003-01-15 Eastern Standard Time" [2] "2003-01-16 Eastern Standard Time" [3] "2003-01-17 Eastern Standard Time" [4] "2003-01-18 Eastern Standard Time" [5] "2003-01-19 Eastern Standard Time" (Actually I'm on Eastern Daylight Time, but let's ignore that for now.) Changing the axes to display not only the date but also the time shows what's going on: f <- "%Y/%m/%d" f2 <- "%Y-%m-%d %H:%M" d <- as.POSIXct(strptime(c("2003/1/15","2003/1/16", "2003/1/17","2003/1/18","2003/1/19"),format=f)) plot(d,1:length(d),axes=F);box() axis(2) abline(v=d[3]) axis.POSIXct(1,d,format=f2) axis.POSIXct(3,at=d,format=f2) The bottom axis put the ticks at 19:00 on each date. This seems an odd time until you note that Eastern Standard Time is 5 hours behind UTC, so 19:00 EST is 00:00 UTC. But is there any easy way to avoid this problem? It is confusing and inconvenient that axis.POSIXct(1,d,format=f) and axis.POSIXct(3,at=d,format=f) give seemingly contradictory scales. Many thanks for any help! Nick Barrowman, Ph.D. Chief Biostatistician, Chalmers Research Group Children's Hospital of Eastern Ontario Research Institute 401 Smyth Road, Ottawa, Ontario, K1H 8L1, Canada Tel (613) 737-7600 ext. 3971 Fax (613) 738-4800 Email [EMAIL PROTECTED] URL www.chalmersresearch.com version _ platform i386-pc-mingw32 arch i386 os mingw32 system i386, mingw32 status major1 minor9.0 year 2004 month04 day 12 language R __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- -- Don MacQueen Environmental Protection Department Lawrence Livermore National Laboratory Livermore, CA, USA __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: RE: [R] ANOVA and contrasts
Dear FZ In one way I have not problem... The problem arise to establish the contrasts for 2 way with interactions and I desire the SS of each one of the contrasts. Thanks, José Cláudio Faria UESC/DCET Brasil 73-634.2779 [EMAIL PROTECTED] [EMAIL PROTECTED] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] any simple examples for SJAVA please
The SJava package is a very flexible and general bidirectional interface between R and Java that can allow not only simple data and complex data structures to be passed between the languages, but also references to objects including using R functions as Java methods. Other approaches to inter-system interfaces are much simpler and direct but as a result are more limited for general computation. There is a great deal of infrastructure under SJava that makes it a general Java interpreter. And building a variety of different communication mechanisms to R really ignores the existng literature in the subject. Unfortunately, there are more opinions about the SJava package than people who actually understand the issues. At the moment, what is a real problem is that there isn't a significant culture for people to help package authors distribute Windows binaries, but more that Windows users require software without any additional effort. Unfortunately, I do not have the time to keep my Windows machine upgraded for service packages, virus software, etc. simply to create binaries for Windows users that I currently do not need. I would be very happy to work with someone who uses Windows and wanted to take on providing binaries. I can both explain and simplify the build process (dropping the support for S-Plus). But it is prohibitive to spend all my time supporting the Omegahat packages and R on numerous platforms while still doing research so that people have fresh ideas in 5 years time. There is a real danger that statistical computing research will become bogged down servicing currrent needs of users and extending concepts that were research topics 20 years ago rather than working on new concepts that will needed for the future. So I am hoping that one or more people can volunteer to help with maintaining the Windows port. It is definitely not trivial as the SJava package is more complex in nature (and implemenation) than many others, but it is feasible. D. Vincent MUTEAUD wrote: > Hi, > like you I'm trying to use SJava and I meet with many difficulties. > I try to run examples from "Calling R from Java" written by Duncan Temple > Lang available from http://www.omegahat.org/SJava (there are many examples > on this web site). > Now I have an error that my R version is R 1.6.1 while I have version 1.9.0 > on windows NT. This causes a crash when I run my Java code: > fatal error: enable to open the base package > I still waiting for a solution > Best > > A 07:58 28/05/2004 -0500, Laura Holt a écrit : > >Hi! > > > >Are there any simple examples for SJAVA please? > > > >Thanks, > >Laura > >R Version 1.9.0 Windows > > > >_ > >Is your PC infected? Get a FREE online computer virus scan from McAfee® > >Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 > > > >__ > >[EMAIL PROTECTED] mailing list > >https://www.stat.math.ethz.ch/mailman/listinfo/r-help > >PLEASE do read the posting guide! > >http://www.R-project.org/posting-guide.html > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! > http://www.R-project.org/posting-guide.html -- Duncan Temple Lang[EMAIL PROTECTED] 371, Kerr Hall University of California at Davis Phone: (530) 752-4782 FAX: (530) 752-7099 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] orca binary?
Uwe Ligges <[EMAIL PROTECTED]> writes: > Laura Holt wrote: > >> Dear R People: >> Is there a binary version for rorca, please? > > > Given I recall correctly, I tried to build rorca for Windows one or > two years ago without success. But maybe Tony (In this context I > recall that I'm still owing him a pizza) knows of a binary? Hah. I should swing by and collect next week. Seriously, you should be able to do a source install or a quick build (I probably could do one later today using the Linux cross-compiler), since it's just raw R code. If someone does it before I do, let me know. best, -tony -- [EMAIL PROTECTED]http://www.analytics.washington.edu/ Biomedical and Health Informatics University of Washington Biostatistics, SCHARP/HVTN Fred Hutchinson Cancer Research Center UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable FHCRC (M/W): 206-667-7025 FAX=206-667-4812 | use Email CONFIDENTIALITY NOTICE: This e-mail message and any attachme...{{dropped}} __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Crossed random effects in lme
Søren Højsgaard <[EMAIL PROTECTED]> writes: > Dear all, > In the SASmixed package there is an example of an analysis of a split-plot > experiment. The model is > > fm1Semi <- lme( resistance ~ ET * position, data = Semiconductor, random = ~ 1 | > Grp) > > where Grp in the Semiconductor dataset is defined as ET*Wafer. Is it possible to > specify the grouping directly some way, e.g. like > > fm1Semi <- lme( resistance ~ ET * position, data = Semiconductor, random = ~ 1 | > ET*Wafer) Actually that does work with lme from the as-yet-unreleased version 0.6-1 of the lme4 package. See the slides from my presentation at useR!2004 http://www.ci.tuwien.ac.at/Conferences/useR-2004/Keynotes/Bates.pdf I'm back at my home now and will endeavor to get lme4_0.6-1 released as soon as possible. -- Douglas Bates[EMAIL PROTECTED] Statistics Department608/262-2598 University of Wisconsin - Madisonhttp://www.stat.wisc.edu/~bates/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Adding key to simple plot() function
I already tried this but encountered a number of problems - firstly the point labels overlie the actual points meaning the labels are unreadable..and I was hoping to have a separate key within which to describe the labels, is this feasible? On Fri, 28 May 2004, Sean Davis wrote: > Laura, > > For your axis issues, look at plotting without axes [plot(,axes=F) and > then see ?axis for making them on your own. As for labeling your data, look > at ?points and ?text. > > Sean > > On 5/28/04 8:27 AM, "Laura Quinn" <[EMAIL PROTECTED]> wrote: > > > I am generating a simple x,y plot to show geographical positions where > > point data has been taken. Rather than plot a graph and lay points on top > > I've opted to generate the map from the point data, ie plot(-x,-y). This > > works fine, though I am wanting to label each point 1:20 and have a key > > alongside - is this possible without getting into trellis plots? I have > > very limited time available so don't want to embark on trellis functions > > which I've never used before.. > > > > Also, my x,y labels currently have a negative sign in front of them which > > I'd like to remove, but if I negate the labels they no longer fit on the > > graph as my data points are -x and -y... > > > > Any suggestions? > > > > Thanks, > > Laura > > > > __ > > [EMAIL PROTECTED] mailing list > > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > > > > __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Merging nlme output
Dear list: I am trying to merge two files together from output I get based on the coef() command. Here is what I am running into. I have two simple linear mixed models > mod1.lme<-lme(math~year, data=sample, random=~year|childid/schoolid) > mod2.lme<-lme(math~year, data=sample, random=~year|childid) I then call the coefficients and store them in the following objects using > mod1.coef<-coef(mod1.lme, level=2) > mod2.coef<-coef(mod2.lme) The problem is that the IDs are different for the two dataframes and I cannot seem to figure out how to make the merge happen nicely. For example, this is output from mod1.lme for the first 5 students > mod1.coef[1:5,] (Intercept) year 2020/273026452 0.03923346 0.9575926 2020/273030991 0.91772318 1.0773731 2020/273059461 0.43865139 1.0111789 2020/278058841 1.11292376 1.0526057 2020/292017571 1.09661340 1.0774692 As opposed to out for the first 5 cases in mod2.coef > mod2.coef (Intercept) year 101480302-1.13483082 0.7023644 173559292 0.52818576 0.8639216 174743401-1.18038537 0.6824612 174755092 0.04760782 0.7402835 179986251-2.42848405 0.5677182 You can see that the ID variable for the first output includes the grouping variable "schoolid" followed by a "/" and then the "childid". I am interested in exploring the differences in the growth trajectories between the two models and would like to merge the two datasets to be able to do so. What do I need to do to somehow eliminate the schoolid and "/" from mod1.lme so that I can use the merge() command? Thanks, Harold C. Doran Director of Assessment One Massachusetts Avenue, NW · Suite 700 Washington, DC 20001-1431 202.336.7075 [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Sorting Data?
On 05/28/04 12:21, Martin Klaffenboeck wrote: >Am 27.05.2004 22:43:02 schrieb(en) Jonathan Baron: >> On 05/27/04 21:58, Martin Klaffenboeck wrote: >> >Hello, >> > >> >Im reading through some manuals, but I cannot find my answer. >> > >> >I have a file containing many data: >> > >> >Vpn CodeFamily Age F1 F2 ... F17 >> >1 1 M 46 1 2 ... 1 >> >2 1 D 18 3 2 ... 4 >> >3 2 M 50 3 3 ... 3 >> >... >> >and so on. >> > >> >Now I can read it by: >> > >> >F = read.table("file", header=T) >> > >> >but now I want to seperate the mothers (M) and daugthers (D) of the >> >family with all the data in all other fields. How can I do that? >> > >> >The 'Code' Tells me which mother belongs to which dougther. I want >> to >> >make a matrix where I have the mothers on one and the daugthers on >> the >> >other axis and compair the distance of every question (F1...F17) and >> >the distance of the sum of this questions. The questions are >> semantic >> >differencials, 5 values. F4, and F7 must have reverse polarity in >> this >> >case. >> >> The following is not tested and probably contains at least one error. > >Thanks, that helps me much as I am a R newbie. > >> Lets assume that there is one mother per daughter and one >> daughter per mother, and your file is Myfile, and the Codes are >> in order. One way is this: > >Ok, we really have only one daughter per mother in our sample. >Im sorting by: > >Myfile <- read.table("Fragebogen.data", header=TRUE) >Myfile <- Myfile[order(e[, 'Code'], Myfile[, 'Family']), ] > >Code has one equal code for mother and daugther the same - so I know >which mother has which daughter, Family tells me if the person it she >mother or the daugther. > >> Myfile$F4 <- -Myfile$F4 # reverse polarity >> Myfile$F7 <- -Myfile$F7 > >Is this also true, if we have a semantic differential with 5 steps? >(from 1 to five. I have one missing value (NA), should I set it to 0?) >(please tell me also if I use incorrect words). >You didn't know that I assume. So now I'm doing: > >Myfile$F3 <- 5-Myfile$F3 > >that seems to be good for me, please tell me what you think. > >> Mothers <- Myfile[Family="M",] >> Daughters <- Myfile[Family="D",] > >Hm. This seems not to work for me i was testing arround, for me seems >to work: > >Mothers <- Myfile[Myfile[["Family"]]=="M",] >Daugthers <- ... My mistake here was to use = instead of ==. If you use my method with ==, it might work too. >I hope we have the same results now. ;-) Im really a newbie in R. > >> Itemdiffs <- Mothers[,-(1:4)]-Daughters[,-(1:4)] # the -(1:4) >> # removes cols 1:4 > >Ok, this seems to work, also but I don't really know what I am doing >with it. Type the names of the variables to find out what you are doing. If they are too big, then subset them, for example, Mothers[1:5,] Daughters[1:5,] Itemdiffs[1:5,] >Also the other things. > >I have to test the hypothesis: Does a daugther answer the questions >(semantic differential) more equal the own mother and more different to >the mothers of the other daugthers. I don't think this is a trivial problem at all, so I am hesitant to offer advice. I see now that you reall do want a matrix, where you have mothers in the columns and daughters in the rows, and distance (difference, similarity) measures in the cells. Perhaps you have several such measures, so you want a three-dimensional array. You might do something like this. First define a function to measure your distance, like Itemdist <- function(x,y) {sum(abs(x[-(1:4)]-y[-(1:4)]))} Dists <- matrix(NA,nrow(Daughters),nrow(Mothers)) for (i in 1:nrow(Daughters) {for (j in 1:nrow(Mothers)) {Dists[i,j] <- Itemdist(Daughters[i,],Mothers[j,])}} Then you would want to show that the diagonal of the resulting matrix is higher (or lower) than the other cells. Here is where I yield to experts. If I were doing it, I might consider comparing these cells to some measure of the expectation of what they ought to be, but I would not just do a t test comparing them to all the other cells because the other cells are not independent of each other. (One daughter might by odd and be dissimilar from everyone, and this would show up in an entire row.) I'm sure there is some simple idea that I'm missing here. >I hope you get that in >english. ;-) It seems that many people on the list read German (including me, but I'm scared to write it), but the official language is English. >Thanks, >Martin > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Jonathan Baron, Professor of Psychology, University of Pennsylvania Home page:http://www.sas.upenn.edu/~baron R search page: http://finzi.psych.upenn.e
Re: [R] any simple examples for SJAVA please
Hi, like you I'm trying to use SJava and I meet with many difficulties. I try to run examples from "Calling R from Java" written by Duncan Temple Lang available from http://www.omegahat.org/SJava (there are many examples on this web site). Now I have an error that my R version is R 1.6.1 while I have version 1.9.0 on windows NT. This causes a crash when I run my Java code: fatal error: enable to open the base package I still waiting for a solution Best A 07:58 28/05/2004 -0500, Laura Holt a écrit : Hi! Are there any simple examples for SJAVA please? Thanks, Laura R Version 1.9.0 Windows _ Is your PC infected? Get a FREE online computer virus scan from McAfee® Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] "privileged slots"
Hi! The simplest way to find the document is ?checkSlotAssignmen A function which is described there and which Matthias Kohl mentioned in his e-mail You will find it also using the html search form of the html documentation searching for RClassUtils. Its true that the document is a little hidden what I havent noticed (Sorry Martin). The functions described in it are not listed in the 00index.html file of the methods package. (Why!?) I found the document by the first trial by typing ?checkSlotAssignment. There was a lot of discussion going on at about S4 at the useR confernce. Much more than at this list. Maybee because the only understandable text about S4 about wich i knew is "Programming with Data" J. Chambers. (The descriptions to the method package is for me quite hard to read.) Maybee also because no one knews how big exactly is the performance sacrifice caused by the additional features. Therefore, this "less" official document is of some value because it allows to make some benchmarks. I was running some test cases to figure out how many(much?) time one can gain by setting check=FALSE check=TRUE. And how using slot instead of @ infuences the execution speed. I run a for loop with x in 1:1 with the following assignments in the loop: [s] [EMAIL PROTECTED] <- paste("test",x,sep="") 2.550 2.71 NA NA slot(tmp,"info",check=T) <- paste("test",x,sep="") 2.360 2.50 NA NA slot(tmp,"info",check=F) <- paste("test",x,sep="") 0.670 0.80 NA NA paste("test",x,sep="") 0.500 0.63 NA NA object.size(tmp) [1] 1208 It means that a type check takes about 0.0002 seconds on a P4 3000 Windows XP with hyperthreding. Its not much. It means thats no reason to be scared of S4 because of performance. The overhad is even less 0.2 Last but not least: There where also some discussions about having the OPTION to pass arguments to function by reference. Sincerely Eryk Ps.: The UseR conference was great! Thanks to the organizers from me too. *** REPLY SEPARATOR *** On 5/27/2004 at 10:55 PM Torsten Steuernagel wrote: >On 27 May 2004 at 19:12, Wolski wrote: > >> But this document is quite intersting. > >Which document are you refering to ? > >> I have just two R questions which arised reading the document which a >> core members of the R-project cant find. > >I can't find it either. > >> 1. Is it intended to introduce privileged slots in the future version of >> R? > >You are talking about access control like C++ (private, protected, >public) has, aren't you ? I can't answer for the R Core Team but I'm >also interested in a similar functionality. Please check the setClass() >docs for the "access" argument. I suppose this is intended to provide >such functionality in the future. > >> 2. Is it true >> that one can gain execution speed by accessing slots using the >> function slots(object,name,check=FALSE)<-value would speed up >> execution? > >Try this: > >> get("@<-") >function (object, name, value) >{ > arg <- substitute(name) > if (is.name(arg)) > name <- as.character(arg) > "slot<-"(object, name, TRUE, value) >} > >So "@<-" actually calls "slot<-". Using "slot<-" instead, you save the >additional overhead the call to "@<-" introduces. If there will be a real >performance gain in replacing "@<-" certainly depends on what you're >doing. > >- Torsten > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] any simple examples for SJAVA please
Hallo Laura! What I have heard at UseR there are no stable running SJava at the Windows platform. ( I wish that I am wrong!) I knew that Simon Urbanek is using and developing a java interface to R. google: simon urbanek Sincerely Eryk *** REPLY SEPARATOR *** On 5/28/2004 at 7:58 AM Laura Holt wrote: >Hi! > >Are there any simple examples for SJAVA please? > >Thanks, >Laura >R Version 1.9.0 Windows > >_ >Is your PC infected? Get a FREE online computer virus scan from McAfee® >Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] any simple examples for SJAVA please
Hi! Are there any simple examples for SJAVA please? Thanks, Laura R Version 1.9.0 Windows _ Is your PC infected? Get a FREE online computer virus scan from McAfee® Security. http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Simple list manipulation question
R> L <- list(A=c("AB","BC","CD"),B=c("GF","HG","FH","FJ")) R> stack(L) values ind 1 AB A 2 BC A 3 CD A 4 GF B 5 HG B 6 FH B 7 FJ B Sean Davis mail.nih.gov> writes: : : I have a list of vectors : : $A : "AB" "BC" "CD" : : $B : "GF" "HG" "FH" "FJ" : : and I want to convert it into a dataframe of form : : A AB : A BC : A CD : B GF : B HG : B FH : B FJ : : Just can't quite come up with a nice "R" solution for it. : : Thanks, : Sean : __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Simple list manipulation question
Hallo! x<-unlist(yourlist) names(x) # maybee you have to postprocess the names a little dataframe(names(x),x) Sincerely Eryk *** REPLY SEPARATOR *** On 5/28/2004 at 8:49 AM Sean Davis wrote: >I have a list of vectors > >$A >"AB" "BC" "CD" > >$B >"GF" "HG" "FH" "FJ" > >and I want to convert it into a dataframe of form > >A AB >A BC >A CD >B GF >B HG >B FH >B FJ > >Just can't quite come up with a nice "R" solution for it. > >Thanks, >Sean > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Simple list manipulation question
I have a list of vectors $A "AB" "BC" "CD" $B "GF" "HG" "FH" "FJ" and I want to convert it into a dataframe of form A AB A BC A CD B GF B HG B FH B FJ Just can't quite come up with a nice "R" solution for it. Thanks, Sean __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
[EMAIL PROTECTED] wrote: > I don't exactly understand what you do, could you show me the > program that you execute to do that? I did such things sometimes ago, so the following is (as usual) without warranty. There are several methods, e.g. using Choleski factorization, singular value decomposition or principal components. Given "mdata" as original data matrix it works with hclust and should be applicable to kmeans too: # with svd z <- svd(scale(mdata, scale=F))$u cl <- hclust(dist(z), method="ward") # with princomp (rescaled) pc <- princomp(mdata, cor=FALSE) pcdata <- as.data.frame(scale(pc$scores)) cl <- hclust(dist(pcdata), method="ward") ... but as I mentioned, this is only an example, that methods working with the Euclidean distance can be applied to other distance measures, when an appropriate transformation of the data exist and, according to Gavin, there are indeed some other possibilities. Thomas P. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] optim(method="SANN")
Hello List I'm working on a combinatoric problem in which the object is to minimize the badness() of a vector. I think this class of problem is only soluble by optim() using method=SANN. The badness() of anything is >= 0, and when I've found a solution with zero badness, I want optim() to stop (carrying on beyond zero badness cannot improve the solution). Efficiency is crucial here. The ?optim manpage states For '"SANN"' 'maxit' gives the total number of function evaluations. There is no other stopping criterion. How best to make optim() stop as soon as it finds a zero badness solution? -- Robin Hankin Uncertainty Analyst Southampton Oceanography Centre SO14 3ZH tel +44(0)23-8059-7743 [EMAIL PROTECTED] (edit in obvious way; spam precaution) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Adding key to simple plot() function
Laura, For your axis issues, look at plotting without axes [plot(,axes=F) and then see ?axis for making them on your own. As for labeling your data, look at ?points and ?text. Sean On 5/28/04 8:27 AM, "Laura Quinn" <[EMAIL PROTECTED]> wrote: > I am generating a simple x,y plot to show geographical positions where > point data has been taken. Rather than plot a graph and lay points on top > I've opted to generate the map from the point data, ie plot(-x,-y). This > works fine, though I am wanting to label each point 1:20 and have a key > alongside - is this possible without getting into trellis plots? I have > very limited time available so don't want to embark on trellis functions > which I've never used before.. > > Also, my x,y labels currently have a negative sign in front of them which > I'd like to remove, but if I negate the labels they no longer fit on the > graph as my data points are -x and -y... > > Any suggestions? > > Thanks, > Laura > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Regression model type II
I am trying to fit regression models type II with R, but it seems to me that most (all) of the linear model functions are for type I regressions. Does anyone knows whether type II regressions functions exist in R. Benjamin -- Benjamin Planque IFREMER Laboratoire d'Ecologie Halieutique BP 21105 44311 Nantes Cedex 03 France Tel: +33 (0)2 40 37 41 17 Fax: +33 (0)2 40 37 40 75 e-mail: [EMAIL PROTECTED] http://www.ifremer.fr/drvecohal/ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Adding key to simple plot() function
I am generating a simple x,y plot to show geographical positions where point data has been taken. Rather than plot a graph and lay points on top I've opted to generate the map from the point data, ie plot(-x,-y). This works fine, though I am wanting to label each point 1:20 and have a key alongside - is this possible without getting into trellis plots? I have very limited time available so don't want to embark on trellis functions which I've never used before.. Also, my x,y labels currently have a negative sign in front of them which I'd like to remove, but if I negate the labels they no longer fit on the graph as my data points are -x and -y... Any suggestions? Thanks, Laura __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] "privileged slots",
On Thu, 27 May 2004 22:55:20 +0200, "Torsten Steuernagel" <[EMAIL PROTECTED]> wrote : >> 2. Is it true >> that one can gain execution speed by accessing slots using the >> function slots(object,name,check=FALSE)<-value would speed up >> execution? > >Try this: > >> get("@<-") >function (object, name, value) >{ >arg <- substitute(name) >if (is.name(arg)) >name <- as.character(arg) >"slot<-"(object, name, TRUE, value) >} > >So "@<-" actually calls "slot<-". Using "slot<-" instead, you save the >additional overhead the call to "@<-" introduces. If there will be a real >performance gain in replacing "@<-" certainly depends on what you're >doing. I'd advise against doing this kind of optimization. It will make your code harder to maintain, and while it might be faster today, if "@<-" is really a major time sink, it's an obvious candidate for optimization in R, e.g. by making it .Internal or .Primitive. When that happens, your "optimized" code will likely be slower (if it even works at all). Duncan Murdoch __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] - making a Windows library from Unix source code
Hi! Normally you are building R packages using cd c:\Program Files\R\rw1081\src R CMD build fatigue Sincerely. *** REPLY SEPARATOR *** On 5/11/2004 at 6:32 PM Livingstone, Paul wrote: >Hi All, > >I'm using R1.8.1 on Windows XP. > >I'm having trouble producing an R library from source code. A colleague >has written the source code, in Unix. I've copied the source code across >to Windows (with the help files, data files, description and index) and am >trying to compile it into a library. > >I've "sourced" each of the *.r files and they appear to work. > >I've very carefully followed the instructions in the "R for Windows FAQ" >and "readme.packages". I've downloaded and installed Perl, R tools, >copied across the text to > >C:\Program Files\R\rw1081\src\library\fatigue > >(note: the package is called "fatigue") then I type at the DOS prompt > >cd "c:\Program Files\R\rw1081\src\gnuwin32" >make pkg-fatigue > >and get the error message(s) > >make[1]: *** [zzzfirst] Error 255 >make: *** [pkg-fatigue] Error 2 > >How do I look up what Error 255 means? >Or can you tell me what Error 255 means and how I might fix it? >Any other hints or ideas on how to use "make" or compile code that appears >to work? > > >thanks, >Paul. > >__ >[EMAIL PROTECTED] mailing list >https://www.stat.math.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Dipl. bio-chem. Eryk Witold Wolski@MPI-Moleculare Genetic Ihnestrasse 63-73 14195 Berlin 'v' tel: 0049-30-83875219 / \ mail: [EMAIL PROTECTED]---W-Whttp://www.molgen.mpg.de/~wolski __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
Gavin Simpson wrote: ... An R example is given in the help file of decostand() in Jari Oksanen's vegan library for two of the transformations mentioned above. ^^^ Pre-empting the usual response about proper terminology, I of course meant package not library. Gav -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [T] +44 (0)20 7679 5522 ENSIS Research Fellow [F] +44 (0)20 7679 7565 ENSIS Ltd. & ECRC [E] [EMAIL PROTECTED] UCL Department of Geography [W] http://www.ucl.ac.uk/~ucfagls/cv/ 26 Bedford Way[W] http://www.ucl.ac.uk/~ucfagls/ London. WC1H 0AP. %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
Thomas Petzoldt wrote: n.bouget wrote: Hi, I want to know which distance is using in the function kmeans and if we can change this distance. Indeed, in the function pam, we can put a distance matrix in parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in the function kmeans, we have to put the matrix of data directly ... Thanks in advance, Nicolas BOUGET One solution is to transform the data in a way, that the euclidean distance of the transformed values represents some other distance of the original values. This works at least for the Mahalanobis-Distance, when one applies a multivariate technique to a PCA transformed and re-scaled matrix, but I don't know if there are transformations for some other distance measures. Thomas P. Other solutions from an ecological paper are: Chord distance Chi square metric Chi square distance Hellinger Distance Distance between species profiles All these can be seen as Euclidean distances of some transformation of the data. The paper "Ecologically meaningful transformations for ordination of species data" Pierre Legendre, and Eugene D. Gallagher (2001) Oecologia Vol. 129, Issue 2, 271-280, explains the concept and how to do the transformations. An R example is given in the help file of decostand() in Jari Oksanen's vegan library for two of the transformations mentioned above. Gav -- %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% Gavin Simpson [T] +44 (0)20 7679 5522 ENSIS Research Fellow [F] +44 (0)20 7679 7565 ENSIS Ltd. & ECRC [E] [EMAIL PROTECTED] UCL Department of Geography [W] http://www.ucl.ac.uk/~ucfagls/cv/ 26 Bedford Way[W] http://www.ucl.ac.uk/~ucfagls/ London. WC1H 0AP. %~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~%~% __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R Tshirts, mugs etc ....
Hi All, I was just wondering if anyone has taken this up. If not, I'd like to volunteer our marketing guy to have a go at sorting this out (he's pretty good with this stuff). Any ideas for designs can be sent directly to him (he's cc'd on this email). When we get some designs and prices, we'll stick them up on a web page so that people can have a look. Btw - Know it's been said a few times already, but ... Great User Conference in Vienna - many thanks to the organizers :-) Cheers, Rich. Mango Solutions Tel +44 (0)1628 418134 Mob +44 (0)7967 808091 -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Andrew Robinson Sent: 26 May 2004 17:41 To: Dirk Eddelbuettel Cc: Achim Zeileis; [EMAIL PROTECTED]; Chad Shaw Subject: Re: [R] thanks again http://www.cafepress.com/ is one such option. I have no personal experience with it. Andrew On Wed, 26 May 2004, Dirk Eddelbuettel wrote: > On Wed, May 26, 2004 at 06:30:40PM +0200, Achim Zeileis wrote: > > As this useR! is over, I guess it's a bit too late for a useR! 2004 > > shirt. > > > > In general, I would agree with you that it would be nice (and not only > > for fun) to have shirts (and coffee mugs and basecaps and ...) with R > > logos or maybe useR! logos. This has been discussed now and then and if > > I recall it correctly the reason that nobody actually started doing it > > is that you would have to spend some time setting it up - and most > > people prefer writing R code instead of mailing R shirts around the > > world. Time was the main reason for me not do organize shirts for the > > useR! - there were so many other things to do and prepare. > > Another point which always kept me from thinking about something like > > this more seriously is the poor quality of the R logo. > > But John Fox mentioned in a discussion that instead of some R user > > doing the work, maybe there are T-shirt mailorders around which could do > > most of the work. So maybe I will have the time to look at this. > > I visited such a site a few months back, but cannot recall the name or URL. > They had e.g. numerous Debian things. If the useR png image was made > available, they could do the commercialisation. With a bit of luck, we may > find a shop that would also donate back a percentage of the proceeds the > family^H^H^H^H^H^foundation. There is probably less merit in the purely > commerical play. > > Dirk > > -- > The relationship between the computed price and reality is as yet unknown. > -- From the pac(8) manual page > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > Andrew Robinson Ph: 208 885 7115 Department of Forest Resources Fa: 208 885 6226 University of Idaho E : [EMAIL PROTECTED] PO Box 441133W : http://www.uidaho.edu/~andrewr Moscow ID 83843 Or: http://www.biometrics.uidaho.edu No statement above necessarily represents my employer's opinion. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html --- Incoming mail is certified Virus Free. [[alternative HTML version deleted]] __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] useR! 2004 keynote lecture slides
Dear useRs, there has been some interest on this list into material about the presentations at useR! 2004. The slides from the keynote lectures are now available from the conference web page at http://www.ci.tuwien.ac.at/Conferences/useR-2004/ thanks to the R-core team members who were willing to give the talks and provide their pdf-slides. For the user-contributed presentations, there are (mostly) only abstracts available on the conference web page - for further information you could look at the web pages of the presenters which are in many cases linked from the participants page. For the useR! organization team, Achim __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
I don't exactly understand what you do, could you show me the program that you execute to do that? > n.bouget wrote: > > Hi, > > I want to know which distance is using in the function kmeans > > and if we can change this distance. > > Indeed, in the function pam, we can put a distance matrix in > > parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but > > we can't do it in the function kmeans, we have to put the > > matrix of data directly ... > > Thanks in advance, > > Nicolas BOUGET > > One solution is to transform the data in a way, that the euclidean > distance of the transformed values represents some other distance of the > original values. This works at least for the Mahalanobis-Distance, when > one applies a multivariate technique to a PCA transformed and re-scaled > matrix, but I don't know if there are transformations for some other > distance measures. > > Thomas P. > Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34/mn) ; tél : 08 92 68 13 50 (0,34/mn) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
n.bouget wrote: Hi, I want to know which distance is using in the function kmeans and if we can change this distance. Indeed, in the function pam, we can put a distance matrix in parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in the function kmeans, we have to put the matrix of data directly ... Thanks in advance, Nicolas BOUGET One solution is to transform the data in a way, that the euclidean distance of the transformed values represents some other distance of the original values. This works at least for the Mahalanobis-Distance, when one applies a multivariate technique to a PCA transformed and re-scaled matrix, but I don't know if there are transformations for some other distance measures. Thomas P. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Sorting Data?
Am 27.05.2004 22:43:02 schrieb(en) Jonathan Baron: On 05/27/04 21:58, Martin Klaffenboeck wrote: >Hello, > >Im reading through some manuals, but I cannot find my answer. > >I have a file containing many data: > >Vpn CodeFamily Age F1 F2 ... F17 >1 1 M 46 1 2 ... 1 >2 1 D 18 3 2 ... 4 >3 2 M 50 3 3 ... 3 >... >and so on. > >Now I can read it by: > >F = read.table("file", header=T) > >but now I want to seperate the mothers (M) and daugthers (D) of the >family with all the data in all other fields. How can I do that? > >The 'Code' Tells me which mother belongs to which dougther. I want to >make a matrix where I have the mothers on one and the daugthers on the >other axis and compair the distance of every question (F1...F17) and >the distance of the sum of this questions. The questions are semantic >differencials, 5 values. F4, and F7 must have reverse polarity in this >case. The following is not tested and probably contains at least one error. Thanks, that helps me much as I am a R newbie. Lets assume that there is one mother per daughter and one daughter per mother, and your file is Myfile, and the Codes are in order. One way is this: Ok, we really have only one daughter per mother in our sample. Im sorting by: Myfile <- read.table("Fragebogen.data", header=TRUE) Myfile <- Myfile[order(e[, 'Code'], Myfile[, 'Family']), ] Code has one equal code for mother and daugther the same - so I know which mother has which daughter, Family tells me if the person it she mother or the daugther. Myfile$F4 <- -Myfile$F4 # reverse polarity Myfile$F7 <- -Myfile$F7 Is this also true, if we have a semantic differential with 5 steps? (from 1 to five. I have one missing value (NA), should I set it to 0?) (please tell me also if I use incorrect words). You didn't know that I assume. So now I'm doing: Myfile$F3 <- 5-Myfile$F3 that seems to be good for me, please tell me what you think. Mothers <- Myfile[Family="M",] Daughters <- Myfile[Family="D",] Hm. This seems not to work for me i was testing arround, for me seems to work: Mothers <- Myfile[Myfile[["Family"]]=="M",] Daugthers <- ... I hope we have the same results now. ;-) Im really a newbie in R. Itemdiffs <- Mothers[,-(1:4)]-Daughters[,-(1:4)] # the -(1:4) # removes cols 1:4 Ok, this seems to work, also but I don't really know what I am doing with it. Also the other things. I have to test the hypothesis: Does a daugther answer the questions (semantic differential) more equal the own mother and more different to the mothers of the other daugthers. I hope you get that in english. ;-) Thanks, Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
On Fri, 28 May 2004, Martin Maechler wrote: > > "n\" == n\ bouget > > on Fri, 28 May 2004 09:37:35 +0200 writes: > > n\> Hi, I want to know which distance is using in the > n\> function kmeans and if we can change this distance. > n\> Indeed, in the function pam, we can put a distance > n\> matrix in parameter (by the line > n\> "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in > n\> the function kmeans, we have to put the matrix of data > n\> directly ... Thanks in advance, Nicolas BOUGET > > It might be interesting to look at this from the pam() > perspective: > What exactly is pam() lacking that kmeans() does for you? > > Christian, are you suggesting that pam() could do the job if > > 1) there was a dist(., method="a la kmeans") > 2) pam() allowed to be started by a user-specified set of >medoids instead of the "Kaufman-Rousseeuw-optimal" ones > ? The k-means criterion is equivalent to: Find a partition C=C_1 \cup...\cup C_k such that \sum_{i=1}^k \sum_{x_j,x_l\in C_i} d(x_j,x_l)/|C_i|=min! d is squared Euklidean distance (see the Bock book). You may wonder to what clustering this would lead with another distance. The difference to pam is that pam minimizes sums of distances to centroid objects, which have to be part of the dataset. k-means does not need centroid objects, no "mean objects" are needed. Thus, pam with squared Euklidean distances is a kind of approximation to k-means. (In practice, both are approximations to a global optimum.) There would also be a further version if other distances would be allowed, the pam criterion would be optimized, but the cluster centers would be allowed to lie elsewhere than on an object of the sample. Of course, pam and the original k-means are more or less easy to compute, while the suggested alternatives may be computationally complex. Best, Christian > > Regards, > Martin Maechler > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > *** Christian Hennig Fachbereich Mathematik-SPST/ZMS, Universitaet Hamburg [EMAIL PROTECTED], http://www.math.uni-hamburg.de/home/hennig/ ### ich empfehle www.boag-online.de __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
> "n\" == n\ bouget > on Fri, 28 May 2004 09:37:35 +0200 writes: n\> Hi, I want to know which distance is using in the n\> function kmeans and if we can change this distance. n\> Indeed, in the function pam, we can put a distance n\> matrix in parameter (by the line n\> "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in n\> the function kmeans, we have to put the matrix of data n\> directly ... Thanks in advance, Nicolas BOUGET It might be interesting to look at this from the pam() perspective: What exactly is pam() lacking that kmeans() does for you? Christian, are you suggesting that pam() could do the job if 1) there was a dist(., method="a la kmeans") 2) pam() allowed to be started by a user-specified set of medoids instead of the "Kaufman-Rousseeuw-optimal" ones ? Regards, Martin Maechler __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] Pr(>|z|) in lme4
Martin, > s <- summary(glmmML( whatever)) > s ## to give the above > str(s) ## to show the internal structure str() is my favorite, it's essentially how I learned R. It was no problem to find my way in nlme, but still lost in lme4. Uwe Liggges wrote: >Well, I wonder what Dieter Menne is looking for. GLMM() is in lme4 and uses S4 classes. glmmML() is in "glmmML" and uses S3 classes.>> You have reasons to wonder. I meant > s <- summary(GLMM( whatever)) but had also used Goran's glmmML,which did not handle na.omit as nicely as GLMM does. Dieter __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Pr(>|z|) in lme4
Martin Maechler wrote: "Dieter" == Dieter Menne <[EMAIL PROTECTED]> on Fri, 28 May 2004 11:04:13 +0200 writes: Dieter> Dear List, I am struggling understanding S4 Dieter> classes. For example, when GLMM Dieter> summary(glmmML( whatever)) Dieter> outputs the following line: Dieter> Estimate Std. Error DF z value Pr(>|z|) Dieter> (Intercept) 0.856 0.319 45 2.68 0.0073 Dieter> How do I access the Pr column? Did you try s <- summary(glmmML( whatever)) s ## to give the above str(s) ## to show the internal structure If 's' itself is an object with a formal class (aka "S4 class"), str() will give sub-optimal (but still helpful) output in the current version of R [exposing the current implementation of S4 slots, which you should never make use of]. Rather in "R Version 2.0.0 Under development (unstable)" aka 'R-devel', it gives much nicer output Martin Well, I wonder what Dieter Menne is looking for. GLMM() is in lme4 and uses S4 classes. glmmML() is in "glmmML" and uses S3 classes. In the latter case, the p-values are calculated within print.glmmML() and the corresponding code is: 1 - pchisq((coef(x)/x$sd)^2, 1) where x is the glmmML object. Uwe Ligges __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Pr(>|z|) in lme4
> "Dieter" == Dieter Menne <[EMAIL PROTECTED]> > on Fri, 28 May 2004 11:04:13 +0200 writes: Dieter> Dear List, I am struggling understanding S4 Dieter> classes. For example, when GLMM Dieter> summary(glmmML( whatever)) Dieter> outputs the following line: Dieter> Estimate Std. Error DF z value Pr(>|z|) Dieter> (Intercept) 0.856 0.319 45 2.68 0.0073 Dieter> How do I access the Pr column? Did you try s <- summary(glmmML( whatever)) s ## to give the above str(s) ## to show the internal structure If 's' itself is an object with a formal class (aka "S4 class"), str() will give sub-optimal (but still helpful) output in the current version of R [exposing the current implementation of S4 slots, which you should never make use of]. Rather in "R Version 2.0.0 Under development (unstable)" aka 'R-devel', it gives much nicer output Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
[EMAIL PROTECTED] wrote: n.bouget wrote: Hi, I want to know which distance is using in the function kmeans and if we can change this distance. Indeed, in the function pam, we can put a distance matrix in parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in the function kmeans, we have to put the matrix of data directly ... Yes but how can we choose the distance to calculate centers? Ah, you are going to use different distance measure (e.g. euclidean, manhattan, ...) as in other cluster methods? Well, that's not possible with the kmeans() implementation. See ?kmeans which tells you: The data given by x is clustered by the k-means algorithm. When this terminates, all cluster centres are at the mean of their Voronoi sets (the set of data points which are nearest to the cluster centre). The algorithm of Hartigan and Wong (1979) is used. Of course, you can do some projection based on the calculation of distances, but I don't think there are functions available to do that completely automatical - and interpretation of results won't be that easy ... Uwe Ligges Thanks in advance, Nicolas BOUGET As the name says, kmeans() calculates *means* (centres) of clusters. It does not any make sense to do that on distances ... Uwe Ligges Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34€/mn) ; tél : 08 92 68 13 50 (0,34€/mn) __ [EMAIL PROTECTED] mailing list https://www.stat. math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34€/mn) ; tél : 08 92 68 13 50 (0,34€/mn) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Statistics Avec R (Vincente Zoonekynd): portuguese translation
On Fri, 2004-05-28 at 00:36, José Cláudio Faria wrote: > Dears friends, > > I have been translating and extended, with educational porposes only, the available > material in the www > (http://zoonek2.free.fr/UNIX/48_R/all.html) "Statistcs avec R" for the Portuguese of > Brazil. > > I consider the material very good and would like to share this work with anothers > users R. > > Already I translated and I extended two chapters: Introduction to the R and analysis > of variance. > > I request to the author (Vincente Zoonekynd) an explicit authorization, a time that > it authorizes it for the English, > but still I did not receive answers. > > I can disponibilizar the material in HTML for the interested parties. > > What do you think about it? > > Best regards, > > José Cláudio Faria > UESC/DCET > Brasil > 73-634.2779 > [EMAIL PROTECTED] > [EMAIL PROTECTED] > Olá, Under what license is this book ? You have to check if the book is licensed and act in agreement with the license rules. You may not need his authorization. Best regards and hope you finish you're translation. I'll be glad to read it. EJ __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Pr(>|z|) in lme4
Dear List, I am struggling understanding S4 classes. For example, when GLMM summary(glmmML( whatever)) outputs the following line: Estimate Std. Error DF z value Pr(>|z|) (Intercept)0.856 0.319 452.68 0.0073 How do I access the Pr column? Dieter __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
RE: [R] R CMD check, latex
It was an (for me not understandable) problem on miktex I´m solving my problem, by copying ..\texmf\miktex\bin\etex.exe to latex.exe I´m wondering that now R CMD check works now. Sorry, for my question to R help. Matthias > -Ursprüngliche Nachricht- > Von: TEMPL Matthias > Gesendet: Freitag, 28. Mai 2004 10:34 > An: [EMAIL PROTECTED] > Betreff: [R] R CMD check, latex > > > Hello, > > I'm trying to build a package and the R CMD build works fine. > > The R CMD check produces an error, because no latex is found. > But I have MikTeX installed and it works fine (only?) in > combination with WinEdt. > > After reinstallation of MikTeX (no errors) I have the same > problem. I´m not sure if this is a problem of my latex or a > problem of the communication from R to my latex. > > (I´m running R 1.9.0 on Windows XP) > > I´m really happy if anybody can help me again. > > Thanks, > Matthias > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo> /r-help > PLEASE > do read the posting guide! > http://www.R-project.org/posting-guide.html > __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Sorting Data?
Am 27.05.2004 22:17:04 schrieb(en) Patrick Burns: First comment: "F" is a bad name to use for an object as it masks the commonly used meaning of FALSE. Oh, thanks, I just took F because 'Fragebogen' is such a long variable. But you are right, I will change that. You'll probably get some better answers, but you can ensure a standard order with: Fsort <- Fdf[order(Fdf[, 'Code'], Fdf[, 'Family']), ] And then perhaps use "reshape". I will look at this closer. Im now reshaping arround and don't know how I can then create a intercorrelation matrix or a (euklidic) distance matrix. But thanks, Martin Patrick Burns Burns Statistics [EMAIL PROTECTED] +44 (0)20 8525 0696 http://www.burns-stat.com (home of S Poetry and "A Guide for the Unwilling S User") Martin Klaffenboeck wrote: Hello, Im reading through some manuals, but I cannot find my answer. I have a file containing many data: VpnCodeFamilyAgeF1F2...F17 11M4612...1 21D1832...4 32M5033...3 ... and so on. Now I can read it by: F = read.table("file", header=T) but now I want to seperate the mothers (M) and daugthers (D) of the family with all the data in all other fields. How can I do that? The 'Code' Tells me which mother belongs to which dougther. I want to make a matrix where I have the mothers on one and the daugthers on the other axis and compair the distance of every question (F1... F17) and the distance of the sum of this questions. The questions are semantic differencials, 5 values. F4, and F7 must have reverse polarity in this case. Hm. How do I have to do now? Thanks, Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] R CMD check, latex
Hello, I'm trying to build a package and the R CMD build works fine. The R CMD check produces an error, because no latex is found. But I have MikTeX installed and it works fine (only?) in combination with WinEdt. After reinstallation of MikTeX (no errors) I have the same problem. I´m not sure if this is a problem of my latex or a problem of the communication from R to my latex. (I´m running R 1.9.0 on Windows XP) I´m really happy if anybody can help me again. Thanks, Matthias __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Sorting Data?
Am 27.05.2004 22:11:04 schrieb(en) Berton Gunter: Your question does not make a lot of sense to me. If you only want to compute distances between mothers and THEIR daughters, you cannot have a matrix, because different mothers have different numbers of daughters. Thats true, but we took only one mother and one daugther in our sample, so in this sample we have only on daugther from one mother. Do you want to define the distance between a mother and someone else's daughter as 0 or INF ? Do you want to show that mothers and their daughters are more similar then mothers and someone else's daughters? Thats my question, yes. ;-) You either need to think about what you wish to do more carefully or express it more carefully. At least for me. Maybe someone else will understand you better. Sorry, its the first time for me to talking about statistics in english and so I have to learn to express it better. Martin -- Bert Martin Klaffenboeck wrote: > Hello, > > Im reading through some manuals, but I cannot find my answer. > > I have a file containing many data: > > Vpn CodeFamily Age F1 F2 ... F17 > 1 1 M 46 1 2 ... 1 > 2 1 D 18 3 2 ... 4 > 3 2 M 50 3 3 ... 3 > ... > and so on. > > Now I can read it by: > > F = read.table("file", header=T) > > but now I want to seperate the mothers (M) and daugthers (D) of the > family with all the data in all other fields. How can I do that? > > The 'Code' Tells me which mother belongs to which dougther. I want to > make a matrix where I have the mothers on one and the daugthers on the > other axis and compair the distance of every question (F1...F17) and > the distance of the sum of this questions. The questions are semantic > differencials, 5 values. F4, and F7 must have reverse polarity in this > case. > > Hm. How do I have to do now? > > Thanks, > Martin > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html -- Bert Gunter Non-Clinical Biostatistics Genentech MS: 240B Phone: 650-467-7374 "The business of the statistician is to catalyze the scientific learning process." -- George E.P. Box __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
> n.bouget wrote: > > > Hi, > > I want to know which distance is using in the function kmeans > > and if we can change this distance. > > Indeed, in the function pam, we can put a distance matrix in > > parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but > > we can't do it in the function kmeans, we have to put the > > matrix of data directly ... Yes but how can we choose the distance to calculate centers? > > Thanks in advance, > > Nicolas BOUGET > > As the name says, kmeans() calculates *means* (centres) of clusters. It > does not any make sense to do that on distances ... > > Uwe Ligges > > > > Accédez au courrier électronique de La Poste : www.laposte.net ; > > 3615 LAPOSTENET (0,34/mn) ; tél : 08 92 68 13 50 (0,34/mn) > > > > __ > > [EMAIL PROTECTED] mailing list > > https://www.stat. math.ethz.ch/mailman/listinfo/r-help > > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html > > Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34/mn) ; tél : 08 92 68 13 50 (0,34/mn) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] gauss.hermite?
On Fri, May 28, 2004 at 09:04:53AM +0200, Dimitris Rizopoulos wrote: > Dear Spencer, > > In Dr. Gray's course notes for Advanced Statistical Computing > (Appendix of Chapter 6) there are some functions for computing > abscissas and corresponding weights for several Gaussian integration > rules: > > http://icommons.harvard.edu/~hsph-bio248-01/Lecture_Notes/ Also have a look at @Book{judd98, author = {Judd, Kenneth L}, title ={Numerical methods in economics}, publisher ={MIT Press}, year = 1998 } especially one of the early chapters ("integration" is in its title, but I don't have the book at the moment), which covers theory and practice of numerical integration and quadrature rules. Implementing these in R has been on my TODO list for a while, so I would be happy to cooperate on this (but only in July or later). Best, Tamas -- Tamás K. Papp E-mail: [EMAIL PROTECTED] Please try to send only (latin-2) plain text, not HTML or other garbage. __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
On Fri, 28 May 2004, Uwe Ligges wrote: > n.bouget wrote: > > > Hi, > > I want to know which distance is using in the function kmeans > > and if we can change this distance. > > Indeed, in the function pam, we can put a distance matrix in > > parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but > > we can't do it in the function kmeans, we have to put the > > matrix of data directly ... > > Thanks in advance, > > Nicolas BOUGET > > As the name says, kmeans() calculates *means* (centres) of clusters. It > does not any make sense to do that on distances ... > > Uwe Ligges That's not really true. There is an equivalent to the k-means target criterion in terms of distances, and that uses squared Euklidean distances. However, as far as I know, you cannot compute it directly in R for any other distance. Using pam is the thing which comes closest. Christian Hennig *** Christian Hennig Fachbereich Mathematik-SPST/ZMS, Universitaet Hamburg [EMAIL PROTECTED], http://www.math.uni-hamburg.de/home/hennig/ ### ich empfehle www.boag-online.de __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] orca binary?
Laura Holt wrote: Dear R People: Is there a binary version for rorca, please? Given I recall correctly, I tried to build rorca for Windows one or two years ago without success. But maybe Tony (In this context I recall that I'm still owing him a pizza) knows of a binary? Uwe Ligges Thanks, Laura mailto: [EMAIL PROTECTED] R Windows 1.9.0 __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] distance in the function kmeans
n.bouget wrote: Hi, I want to know which distance is using in the function kmeans and if we can change this distance. Indeed, in the function pam, we can put a distance matrix in parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in the function kmeans, we have to put the matrix of data directly ... Thanks in advance, Nicolas BOUGET As the name says, kmeans() calculates *means* (centres) of clusters. It does not any make sense to do that on distances ... Uwe Ligges Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34€/mn) ; tél : 08 92 68 13 50 (0,34€/mn) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] Building packages and Rcmd check
Laura Holt wrote: Hello yet again! I'm trying to build a package. The Rcmd build works fine. But the Rcmd check produces an error because I don't have latex on my laptop. Any suggestions, please? Rcmd check --help tells you that you can use the option: --no-latexdo not run LaTeX on help files if you don't want LaTeX porcessing to be checked. ... or install LaTeX ... Uwe Ligges R Windows 1.9.0 thanks again, Laura _ Learn to simplify your finances and your life in Streamline Your Life from MSN Money. http://special.msn.com/money/0405streamline.armx __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] "privileged slots"
Martin Maechler schrieb: "Matthias" == Matthias Kohl <[EMAIL PROTECTED]> on Thu, 27 May 2004 14:01:51 +0100 writes: Matthias> Hi all, in the help for RClassUtils I found the Matthias> expression "privileged slots" in function Matthias> "checkSlotAssignment" with the explanation: Matthias> /privileged slots (those that can only be set by Matthias> accesor functions defined along with the class Matthias> itself)/ RClassUtils ??? > help.search("RClassUtils") your right, sorry but, at least a R Site search in "Functions" gives me one match: "Utilities for Managing Class Definitions" which hast the "title": RClassUtils{methods} R Documentation No help files found with alias or concept or title matching 'RClassUtils' using fuzzy matching. - So I guess that's not something in a standard R document. You should rather keep to the 'official documentation' ... I thought this is a official documentation ... Matthias> I thought all slots of a (not private) class can Matthias> be a accessed by a user via the @ Operator. I tend to agree with your thoughts... Matthias> Is there a way to make a single slot of a class (not Matthias> the whole class) private, so that you can access Matthias> this slot only via an accessor function (not via @)? I'd rather guess not. Matthias> Thanks, for your help Matthias Martin __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] Building packages and Rcmd check
Hello yet again! I'm trying to build a package. The Rcmd build works fine. But the Rcmd check produces an error because I don't have latex on my laptop. Any suggestions, please? R Windows 1.9.0 thanks again, Laura _ Learn to simplify your finances and your life in Streamline Your Life from MSN Money. http://special.msn.com/money/0405streamline.armx __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
[R] distance in the function kmeans
Hi, I want to know which distance is using in the function kmeans and if we can change this distance. Indeed, in the function pam, we can put a distance matrix in parameter (by the line "pam<-pam(dist(matrixdata),k=7)" ) but we can't do it in the function kmeans, we have to put the matrix of data directly ... Thanks in advance, Nicolas BOUGET Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34/mn) ; tél : 08 92 68 13 50 (0,34/mn) __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Re: [R] gauss.hermite?
Dear Spencer, In Dr. Gray's course notes for Advanced Statistical Computing (Appendix of Chapter 6) there are some functions for computing abscissas and corresponding weights for several Gaussian integration rules: http://icommons.harvard.edu/~hsph-bio248-01/Lecture_Notes/ I hope this helps. Best, Dimitris Dimitris Rizopoulos Doctoral Student Biostatistical Centre School of Public Health Catholic University of Leuven Address: Kapucijnenvoer 35, Leuven, Belgium Tel: +32/16/396887 Fax: +32/16/337015 Web: http://www.med.kuleuven.ac.be/biostat/ http://www.student.kuleuven.ac.be/~m0390867/dimitris.htm - Original Message - From: "Spencer Graves" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Friday, May 28, 2004 6:35 AM Subject: [R] gauss.hermite? > The search at www.r-project.org mentioned a function > "gauss.hermite{rmutil}". However, 'install.packages("rmutil")' > produced, 'No package "rmutil" on CRAN.' How can I find the current > status of "gauss.hermite" and "rmutil"? > > Thanks, > Spencer Graves > > __ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html __ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html