similar to: Bug in qr.R ? (PR#9655)

Hi there, Given a positive definite symmetric matrix, I can use chol(x) to obtain U where U is upper triangular and x=U'U. For example, x=matrix(c(5,1,2,1,3,1,2,1,4),3,3) U=chol(x) U # [,1] [,2] [,3] #[1,] 2.236068 0.4472136 0.8944272 #[2,] 0.000000 1.6733201 0.3585686 #[3,] 0.000000 0.0000000 1.7525492 t(U)%*%U # this is exactly x Does anyone know how to obtain L such

The following snippet suggests that there is either a bug in qr(,LAPACK=T), or some bug in my understanding. Note that the detected rank is correct (= 2) using the default LINPACK qr, but incorrect (=3) using LAPACK. This is running on Linux Redhat 9.0, using the lapack library that comes with the Redhat distribution. I'm running R 1.7.1 compiled from the source. If the bug is in my

Dear All i have a little puzzle about eigenvector in the R. As we know that the eigenvector can be displayed on several form. For example A=matrix(c(1,2,4,3),2,2) if we want to get the eigenvalue and eigenvector, the code followed eigen(A) $values [1] 5 -1 $vectors [,1] [,2] [1,] -0.7071068 -0.8944272 [2,] -0.7071068 0.4472136 however, we also can calculate the vector matrix

hi all i have a question that about the eigen analysis found in R and in eviews. i used the same data set in the two packages and found different answers. which is incorrect? the data is: aa ( a correlation matrix) 1 0.9801 0.9801 0.9801 0.9801 0.9801 1 0.9801 0.9801 0.9801 0.9801 0.9801 1 0.9801 0.9801 0.9801 0.9801 0.9801 1 0.9801 0.9801 0.9801 0.9801 0.9801 1 now > svd(aa) $d [1] 4.9204

Hi, All: What tools exist for diagnosing singular gradient problems with 'nls'? Consider the following toy example: DF1 <- data.frame(y=1:9, one=rep(1,9)) nlsToyProblem <- nls(y~(a+2*b)*one, DF1, start=list(a=1, b=1), control=nls.control(warnOnly=TRUE)) Error in nlsModel(formula, mf, start, wts) : singular gradient matrix at initial

Hi, with small matrices eigen works as expected: > eigen(cbind(c(1,4),c(4,7)), only.values = TRUE) $values [1] 9 -1 $vectors NULL > eigen(cbind(c(1,4),c(4,7))) $values [1] 9 -1 $vectors [,1] [,2] [1,] 0.4472136 -0.8944272 [2,] 0.8944272 0.4472136 > eigen(cbind(c(1,-1),c(1,-1))) $values [1] -3.25177e-17+1.570092e-16i -3.25177e-17-1.570092e-16i $vectors

Dear HelpeRs, I wonder if anyone knows of ways to read EViews file types. I did not find a function in the package 'foreign' and a search query submitted to http://search.r-project.org was not successful. Any hint is very much welcome. Dietrich Trenkler -- Dietrich Trenkler c/o Universitaet Osnabrueck Rolandstr. 8; D-49069 Osnabrueck, Germany email: Dietrich.Trenkler at

Dear HelpeRs, I'm very fond of Sweave and I use it as often as possible. It'a a pity I can't use it for larger projects or can I? For instance suppose I have three files file1.rnw, file2.rnw and file3.rnw with Sweave code. Working on file2.rnw I whould like to exclude file1.rnw and file3.rnw temporarily and joining all of them later. This amounts to a mechanism similar to using

Niels Waller wrote: > Does anyone know of an R (or S-PLUS) function for delineating all possible > combinations and permutations? The following function delivers all permutations of 1:n. all.perm <- function(n) { p <- matrix(1, ncol = 1) for (i in 2:n) { p <- pp <- cbind(p, i) v <- c(1:i, 1:(i - 1)) for (j in 2:i) { v <- v[-1]

I have some data which I would like to display with dotchart. The labels are very long, so the chart becomes too small. Setting cex=0.7 seems to be a good compromise, but the gap between the text and the chart still is too large. I did not find a "gap" parameter in the description of dotchart... Thanks for any help. D. Trenkler "a" <- structure(c(103.35, 36.73, 55.09,

Dear HelpeRs, I have some data: "ice" <- structure(c(0.386, 0.374, 0.393, 0.425, 0.406, 0.344, 0.327, 0.288, 0.269, 0.256, 0.286, 0.298, 0.329, 0.318, 0.381, 0.381, 0.47, 0.443, 0.386, 0.342, 0.319, 0.307, 0.284, 0.326, 0.309, 0.359, 0.376, 0.416, 0.437, 0.548, 41, 56, 63, 68, 69, 65, 61, 47, 32, 24, 28, 26, 32, 40, 55, 63, 72, 72, 67, 60, 44, 40, 32, 27, 28, 33,

I think I've discovered what went wrong. My workspace included a function wilcox.test formerly copied from ctest. Now ctest ist part of the distribution and because of that I always got the message: [Previously saved workspace restored] Error in autoload("wilcox.test", "ctest") : Object already exists I didn't take care of this message. Now I removed

Hallo everyone, excuse me if this is not a genuine R question but I do not know where to ask else. Referring to e.g. https://stat.ethz.ch/pipermail/r-help/2004-December/062114.html I wonder if these measurements of 3000 criminals (raw data) are available anywhere. At least I didn't find them in the R datasets package or by means of Google. What I did find was a table of frequencies of

I suspect that there is bug in sub when using "?": > string_"This is a bug!" > sub("!", ", or isn't it?", string) [1] "This is a bug, or isn't it?" > string_"This is a bug?" > sub("?", ", or isn't it?", string) [1] "This is a bug?" Regards, *** D.Trenkler ***

I have two lists of equal lengths. Each one consists of vectors. Is there an elegant way to combine them elementwise? Elegant means: without loops. For instance, for > jja_list(1:3,4:7) > jjb_list(rep(0,4),(2,5)) I would like to get: > list(c(1:3,rep(0,4)),c(4:7,rep(2,5))). --- D.Trenkler --- ************************************************************************ *********

Does anyone know of a way of dealing with unbalanced mixed effects (fixed and random factors) for fully factorial designs. An example of such data is given below; The response variable is SQRTRECRUITS SEASON is a random factor DENSITY is a fixed factor Thus DENSITY:SEASON is a fixed factor. Therefore, whereas the effects of SEASON and DENSITY:SEASON should be tested against the overall

Excuse me if this question has been answered before. I just came across "Writing R documentation" in the "Writing R extensions" manual. Under section 2.9: "Processing Rd format" I only find hints at generating the Rd format on a Unix platform. Have I overlooked how to generate them under Windows? Any help is gratefully appreciated. --- D.Trenkler ---

I'm an old APL fossil and I got used to work with nested arrays there. In APL(2) you can write 1+(1 2 3)(4 5 6 7 8 9) and get (2 3 4)(5 6 7 8 9 10). A similar concept is R's lists. So it takes me by surprise to learn that 1+list(1:3,4:9) delivers an Error in 1 + list(1:3, 4:9) : non-numeric argument to binary operator Is there any particular reason why this doesn't work

For some small sample sizes I would like to exemplify the combinatorics underlying certain nonparametric tests like Mann-Whitney-Wilcoxon, Kruskal-Wallis and Spearman's rank correlation. I have written a function all.perm which delivers all permutations of 1,2,...,n and which works reasonably well. This can be used to generate P(R=r) of Spearman's rank correlation:

Dear helpeRs, a colleague of mine would like to give R a try. He uses econometric models which typically involve a large number of variables, esp. time series. Having no experience with handling very large data sets myself I turn to you. 1. Could you please describe your experiences to cope with these situations? 2. What kind of difficulties will he have to face? Are there special