similar to: couldn't find FUN

My previous post about prcomp and princomp was done in some haste as I had long ago indicated to Kurt that I would try to have this ready for the June release, and it appeared that I would miss yet another release. I also need to get it out before it becomes hopelessly buried by other work. Brian Ripley kindly pointed out some errors, and also pointed out that I was suggesting replacing some

According to R help: princomp() uses eigenvalues of covariance data. prcomp() uses the SVD method. yet when I run the (eg., USArrests) data example and compare with my own "hand-written" versions of PCA I get what looks like the opposite. Example: comparing the variances I see: Using prcomp(USArrests) ------------------------------------- Standard deviations: [1] 83.732400 14.212402

Hi, I would like to use sweep to "sweep out" proportions and confidence intervals for an array, however when I supply a function which returns a string (containing something like "9% (3-18%)") I get back a list instead of an array, here is a simplified example: # example showing that sweep does not return an array with same dimensions as STATS as advertised

I've been working on a revised version of prcomp and princomp. Below is my current draft of prcomp, which is marginally different from V&R. I've added center and scale as optional arguments. However, scale causes the following: > zi _ prcomp(iris[,,2]) Warning: ignored non function "scale" because scale is both a variable and a function. Is there any way to avoid this

Per the discussion about the problems with prcomp() when n << p, which boils down to a problem with svd() when n << p, here is a patch to prcomp() which substitutes La.svd() instead of svd(). -Greg (This is really a feature enhancement, but submitted to R-bugs to make sure it doesn't get lost. ) *** R-1.6.0/src/library/mva/R/prcomp.R Mon Aug 13 17:41:50 2001 ---

In the manual page for prcomp(), it says that sdev is "the standard deviations of the principal components (i.e., the square roots of the eigenvalues of the covariance/correlation matrix, though the calculation is actually done with the singular values of the data matrix)." ?However, this is not what I'm finding. ?The values appear to be the standard deviations of a reprojection of

This continues the message "data.frame within a function (PR#9294)" that was posted on 2006/10/12. Duncan Murdoch kindly replied. I'm using the current version R 2.4.0, but the same issue exists. Just copy and paste the following code under R, and compare the output of f1() and f2() and the output of f3() and f4(). Does anybody have any idea? Thanks.

Lately I'm getting this error quite a bit: Error in La.svd(x, nu, nv) : error code 1 from Lapack routine 'dgesdd' I'm running R 2.5.0 on a 64 bit Intel machine running Fedora (8 I think). Maybe the 64 bit platform is more fragile about declaring convergence. I'm seeing way more of these errors than I ever have before. From R-Help I see that this issue comes up from time to

Something I can't quite place finger on here. Altogether too much Mafia, in the bulk email business generally, and I know Switzerland borders on Italy ... This sounds, (albeit vaguely,) altogether too much like the thieves I seem to have fallen amongst lately. Two stolen trucks, three stolen laptops, another one wrecked, three or four stolen cell phones, passwords GPG keys, city hall hookers

Interrogating some (of my own) code in another package. >norm.meth <- getMethod("normalize", "MatrixLike") >message("str(norm.meth)") >str(norm.meth) >message("show(norm.meth at .Data)") >show(norm.meth at .Data) Last show() displays this: function (object, ...) { .local <- function (object, method = c("median",

I'm attempting to reflect the information for use with corresponding fields in GUI (in a different package), to provide default values, argname as key for UI label lookups, etc. So I want something much more like the formals of the implementation: { "object", "method": c("median", "vs", "tukey"),

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, this is probably simple and I'm just doing something stupid, sorry about that :-) I'm trying to convert words (strings of letters) into a fairly small dimensional space (say 10, but anything between about 5 and 50 would be ok), which I will call a feature vector. The the distance between two words represents the similarity of the

Hi, I'm trying to do PCA on a n by p wide matrix (n < p), and I'd like to get more principal components than there are rows. However, svd() only returns a V matrix of with n columns (instead of p) unless the argument nv=p is set (prcomp calls svd without setting it). Moreover, the eigenvalues returned are always min(n, p) instead of p, even if nv is set: > x <-

Hi, I'm just wondering if it is planned to include the Lapack routine DGESDD (and friends) in R-1.4.0? This is faster (supposedly by a factor of ~6 for large matrices) than DGESVD which is currently (R-1.3.1) called by La.svd. And if it is not in the plans yet, is there a chance it could be? I've added it to my local version of R-1.3.1 and so far see a factor of 4 improvement over

--=====================_24736660==_ Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: quoted-printable SVD-Error on R 1.1.0 Windows 98 I get the following error applying svd on a positive definite matrix : > sk2 [,1] [,2] [,3] [,4] [,5] [1,] 1.0460139783 0.084356992 -2.810553e-04

I am trying to perform Singular Value Decomposition (SVD) on a Term Document Matrix I created using the 'tm' package. Eventually I want to do a Latent Semantic Analysis (LSA). There are 5677 documents with 771 terms (the DTM is 771 x 5677). When I try to do the SVD, it runs out of memory. I am using a 12GB Dual core Machine with Windows XP and don't think I can increase the memory

I am just about working through Faraways excellent tutorial "practical regression and ANOVA using R" on page 24 he makes the x matrix: x <- cbind(1,gala[,-c(1,2)]) how can I understand this gala[,-c(1,2)])... I couldn't find an explanation of such "c-like" abbreviations anywhere. thanks for a hint. another problem: I couldn't load the faraway library, using the

Dear R-devel: I could use some advice about matrix calculations and steps that might make for faster computation of generalized inverses. It appears in some projects there is a bottleneck at the use of svd in calculation of generalized inverses. Here's some Rprof output I need to understand. > summaryRprof("Amelia.out") $by.self self.time self.pct

Hi! Excuse the delay, but I had to deal with a corrupted NTFS file system that ate many important files on an USB stick... The FLAC version of the original is almost 6MB and it can be downloaded slowly from this time-limited link: https://sbr5vjid0jgmce4q.myfritz.net:40262/nas/filelink.lua?id=0ba5a10529a6fe7b On the meaning of a logarithmic sweep: If you use foobar2000 and the

Hi Jean-Marc and others, I put Speex support in Sweep (a sound editor) yesterday, and had some fun with it :) I'd like to get some feedback before releasing it publically. I'm not really used to speech codecs (I work more with music) but I've been quite impressed at the file size and quality (especially when you hear what I did with some of the sample files ;) I didn't have any