similar to: How to generate regression matrix with correlation matrix

Hi All. I'd like to generate a sample of n observations from a k dimensional multivariate normal distribution with a random correlation matrix. My solution: The lower (or upper) triangle of the correlation matrix has n.tri=(d/2)(d+1)-d entries. Take a uniform sample of n.tri possible correlations (runi(n.tr,-.99,.99) Populate a triangle of the matrix with the sampled correlations Mirror the

Dear R community: I am trying to do a simulation study mentioned by Fu (1998), Journal of Computational and Graphical Statistics, Volume7, Number 3, Page 397-416. In order to give a clear statement of quesion I copy the following paragraph from the article: We compare the bridge model with the OLS, the lasso and the ridge in a simulation of a linear regression model of 30 observations and 10

General goal: Write R code to find the inverse matrix of an nxn positive definite symmetric matrix. Use solve() to verify your code works. Started with a 3x3 matrix example to build the code, but something dosen't seem to be working. I just don't know where I am going wrong. ##Example matrix I found online A<-c(4,1,-1,1,2,1,-1,1,2) m<-matrix(A,nrow=3,ncol=3) ##Caculate the eigen

Hi, If a matrix is not positive definite, make.positive.definite() function in corpcor library finds the nearest positive definite matrix by the method proposed by Higham (1988). However, when I deal with correlation matrices whose diagonals have to be 1 by definition, how do I do it? The above-mentioned function seem to mess up the diagonal entries. [I haven't seen this complication, but

Hi, I have a data.frame with 294 columns and 211 rows. I am calculating correlations between all pairs of columns (excluding column 1) and based on these correlation values I delete one column from any pair that shows a R^2 greater than a cuttoff value. (Rather than directly delete the column all I do is store the column number, and do the deletion later) The code I am using is: ndesc

Hi: Does R have a function as gsorth is SAS, that perform a the Gram-Schmidt orthonormal factorization of the m ?n matrix A, where m is greater than or equal to n? That is, the GSORTH subroutine in SAS computes the column-orthonormal m ?n matrix P and the upper triangular n ?n matrix T such that A = P*T. or any other version of Gram-Schmidt orthonormal factorization? I search the help, but I

Hi, Why the qr() produces a negative Q compared with Gram-Schmidt? (note example below, except Q[2,3]) Here is an example, I calculate the Q by Gram-Schmidt process and compare the output with qr.Q() a <- c(1,0,1) b <- c(1,0,0) c <- c(2,1,0) x <- matrix(c(a,b,c),3,3) ########################## # Gram-Schmidt ########################## A <- matrix(a,3,1) q1 <-

Dear all, Sorry to post my query once again in the list, since I did not get attention from anyone in my previous mail to this list. Now I make it simple here that please give me a code for find out the columns of a dataframe whose correlation coefficient is below a pre-determined threshold. (For detailed query please see my previous message to this list, pasted hereunder) Thanks and regards,

Hi, I was trying to evaluate the pearson correlation and the p-values for an nxm matrix, where each row represents a vector. One way to do it would be to iterate through each row, and find its correlation value( and the p-value) with respect to the other rows. Is there some function by which I can use the matrix as input? Ideally, the output would be an nxn matrix, containing the p-values

Hi UseRs, I don't want to die beeing idiot... I dont understand the different results between: cor() and cov2cov(cov()). See this little example: > x=matrix(c(0.5,0.2,0.3,0.1,0.4,NA,0.7,0.2,0.6,0.1,0.4,0.9),ncol=3) > cov2cor(cov(x,use="pairwise.complete.obs")) [,1] [,2] [,3] [1,] 1.0000000 0.4653400 -0.1159542 [2,] 0.4653400 1.0000000

Is there an R function for computing a variance-covariance matrix that guarantees that it will have no negative eigenvalues? In my case, there is a *lot* of missing data, especially for a subset of variables. I think my tactic will be to compute cor(x, use="pairwise.complete.obs") and then pre- and post-multiply by a diagonal matrix of standard deviations that were computed based

Hi. I have been looking at the PCAproj function in package pcaPP (R 2.4.1) for robust principal components, and I'm trying to interpret the results. I started with a data matrix of dimensions RxC (R is the number of rows / observations, C the number of columns / variables). PCAproj returns a list of class princomp, similar to the output of the function princomp. In a case where I can

I've a vector of pairwise correlations in the order low-index element precedes the high-index element, say: corr(1,2)=0.1, corr(1,3)=0.2, corr(2,3)=0.3, corr(3,4)=0.4 How can I construct the corresponding correlation matrix? I tried using the "combn"-function in "combinat" package: library(combinat) combn(c(0.1,0.2,0.3,0.4),2) , but to no avail... Thank you for your

Hello List, I would like to orthonormalize vectors contained in a matrix X taking into account row weights (matrix diagonal D). ie, I want to obtain Z=XA with t(Z)%*%D%*%Z=diag(1) I can do the Gram-Schmidt orthogonalization with subsequent weighted regressions. I know that in the case of uniform weights, qr can do the trick. I wonder if there is a way to do it in the case of non uniform

Dear R list users, the singluar value decomposition of a symmetric matrix M is UDV^(T), where U = V. La.svd(M) gives as output three elements: the diagonal of D and the two orthogonal matrices u and vt (which is already the transpose of v). I noticed that the transpose of vt is not exactly u. Why is that? thank you for your attention and your help Stefano AVVISO IMPORTANTE: Questo messaggio di

List members, Via searches I've seen similar discussion of this topic but have not seen resolution of the particular issue I am experiencing. If my search on this topic failed, I apologize for the redundancy. I am attempting to generate random covariance matrices but would like the corresponding correlations to be uniformly distributed between -1 and 1. The approach I have been using is:

It is not clear to me that one can. If the singular value decomposition of A is the triple product P d Q', then the singular value decomposition of A'A=S is Q d^2 Q'. The information about the orthonormal matrix P is lost, is it not? ********************************************************** Cliff Lunneborg, Professor Emeritus, Statistics & Psychology, University of Washington,

Dear listers, I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. Everything went OK except this: sudo R CMD check pgirmess ..... * checking examples ... ERROR Running examples in 'pgirmess-Ex.R' failed. The error most

I have to admit that I'm at a bit of a loss here; any pointers would be greatly appreciated. I've been making correlation matricies from some of my datasets, and have been instructed to get the probability values for each of these correlations. I've checked the online help for info on both the cor and cov functions, but I was unable to find any relevant info on finding how to obtain

> I am trying to produce a heatmap of pairwise correlations, but since the matrix is > symmetric, I only need either the upper or the lower triangle. I have scoured the > web and R documentation, but I have not been able to find a way to produce such a > figure. Is there a simple way to produce a heat map with only the part above or > below the diagonal? You might want to check