similar to: matrix help

I am trying to estimate a covariance matrix from the Hessian of a posterior mode. However, this Hessian is indefinite (possibly because of numerical/roundoff issues), and thus, the Cholesky decomposition does not exist. So, I want to use a modified Cholesky algorithm to estimate a Cholesky of a pseudovariance that is reasonably close to the original matrix. I know that there are R packages that

In an experimental setup we obtain z-data samples at equidistant grid points. The surf.gls (Kriging) algorithm produces an error under this circumstance when performing the Choleski decomposition. A workaround is to dither the grid coordinates using (x <- rnorm(length(x)) ; y<- rnowm(length(y))). Question: Is this an expected behaviour of the surf.gls function ? Regards, Berthold

Dear all, I made a simple test of the Cholesky decomposition in the package 'Matrix', by considering 2 variables 100% correlated. http://blogs.sas.com/content/iml/2012/02/08/use-the-cholesky-transformation-to-correlate-and-uncorrelate-variables/ The full code is below and can be simply copy&paste in the R prompt. After uncorrelation I still have a correlation of +-100%...

Hi everyone: I try to use r to do the Cholesky Decomposition,which is A=LDL',so far I only found how to decomposite A in to LL' by using chol(A),the function Cholesky(A) doesnt work,any one know other command to decomposte A in to LDL' My r code is: library(Matrix) A=matrix(c(1,1,1,1,5,5,1,5,14),nrow=3) > chol(A) [,1] [,2] [,3] [1,] 1 1 1 [2,] 0 2 2

Dear R-devel members, I am looking for a fast Cholesky update/downdate. The matrix A being symmetric positive definite (n, n) and factorized as A = L %*% t(L), the goal is to factor the new matrix A +- C %*% t(C) where C is (n, r). For instance, C is 1-column when adding/removing an observation in a linear regression. Of special interest is the case where A is sparse. Looking at the

I get the following error from one of my Rd files in R CMD check (R 2-14.0) * checking Rd files ... WARNING Error in switch(attr(block, "Rd_tag"), TEXT = if (!grepl("^[[:space:]]* $", : EXPR must be a length 1 vector problem found in ?backsolve.Rd? This is likely something that will be glaringly obvious once it's pointed out, but without a line number I can't

Hello, I am updating my benchmark (http://www.sciviews.org/other/benchmark.htm) to recent versions of data analysis software (including R 1.6.0 and Splus 6.1), and I now run it on a Pentium IV instead of the old Celeron 500 Mhz that candidates for retirement. I test R under Windows Xp pro with and without optimized BLAS. I use the optimized Rblas.dll for P4 found on CRAN. Here are the results.

Hi all, This question is not really R related, rather on Statistics subject itself. Even I did not do those using R. however still I want to post it here, because my hope is I could get help from great statisticians who are the very active member of this group. My problem is to interpret Variance decomposition of VAR model in layman's language. Using EViews I got following : Variance

Dear fellow R Users: I am doing a Cholesky decomposition on a correlation matrix and get error message the matrix is not semi-definite. Does anyone know: 1- a work around to this issue? 2- Is there any approach to try and figure out what vector might be co-linear with another in thr Matrix? 3- any way to perturb the data to work around this? Thanks for any suggestions.

A new version of kernlab has just been released. kernlab is a kernel-based Machine Learning package for R. kernlab includes the following functions: o ksvm() : Support Vector Machines for classification, regression, novelty detection, native multi-class classification, support for class-probability output and confidence intervals in regression. o gausspr()

A new version of kernlab has just been released. kernlab is a kernel-based Machine Learning package for R. kernlab includes the following functions: o ksvm() : Support Vector Machines for classification, regression, novelty detection, native multi-class classification, support for class-probability output and confidence intervals in regression. o gausspr()

Dear list, I would like to do something like: # Simulate some example data tmp <- matrix(rnorm(50), ncol=5) colnames(tmp) <- c("y","x1","x2", "x3", "x4", "x5") # Fit a linear model with random noise added to x5 n times n <- 100 replicate(n, lm(y ~ x1+x2+x3+x4+I(x5+rnorm(nrow(tmp))), data=as.data.frame(tmp))) I am wondering

Hi, I want to simulate a data set with similar covariance structure as my observed data, and have calculated a covariance matrix (dimensions 8368*8368). So far I've tried two approaches to simulating data: rmvnorm from the mvtnorm package, and by using the Cholesky decomposition (http://www.cerebralmastication.com/2010/09/cholesk-post-on-correlated-random-normal-generation/). The problem is

2 days ago, I posted my long message about the observed slowdown in a package between R-2.15.0 and R-2.15.2. Uwe Ligges urged me to make a self-contained R example. That was the encouragement I needed. I tracked the problem down to a failing use of a LAPACK routine. R's LAPACK C interface changed one variable in one function. But it turned out to be an important change. In case others have

Hi Paul, We've been through this process ourselves for the Revolution R Open project. There are a number of pitfalls to avoid, but you can take a look at how we achieved it in the build scripts at: https://github.com/RevolutionAnalytics/RRO There are also some very useful notes in the R Installation guide: https://cran.r-project.org/doc/manuals/r-release/R-admin.html#BLAS Most packages do

Hello Everyone I have a MCMC loop to calculate a time varying hierarchical Bayesian structure. This requires me to use around 5-6 matrix inversions in the loop. I use cholesky and chol2inv for the matrix decomposition. Because of the data I am working with I am required to invert a 167 by 167 matrix twice in one iteration. I need to run the iteration for 10000 times, but I get the error

I cleaned up my old benchmarking code and added checks for missing data to compare various ways of finding OLS regression coefficients. I thought I would share this for others. the long and short of it is that I would recommend ols.crossprod = function (y, x) { x <- as.matrix(x) ok <- (!is.na(y))&(!is.na(rowSums(x))) y <- y[ok]; x

I'm trying to use stepAIC on sparse matrices, and I need some help. The documentation for slm.fit suggests: slm.fit and slm.wfit call slm.fit.csr to do Cholesky decomposition and then backsolve to obtain the least squares estimated coefficients. These functions can be called directly if the user is willing to specify the design matrix in matrix.csr form. This is often advantageous in large

I am a relative newcomer to both the R and C/C++ software worlds -- I'm taking a C Programming class currently. I noticed the other day that the C:\Program Files\R1_8_1\src\include\R_ext directory on my WinXP box has the header files BLAS.h Lapack.h Linpack.h RLapack.h I am interested in (perhaps) using one or more of these header files in a straight C program I'm working on in Visual

Dear useRs, I have written a function that implements a Bayesian method to compare a patient's score on two tasks with that of a small control group, as described in Crawford, J. and Garthwaite, P. (2007). Comparison of a single case to a control or normative sample in neuropsychology: Development of a bayesian approach. Cognitive Neuropsychology, 24(4):343?372. The function (see