similar to: Multivariate regression in R

To estimate the covariance matrix of e you could use the sample covariance matrix of the residuals. If desired, use its cholesky decomposition to transform to make the error approximately uncorrelated, then refit (and back-transform the coefficient matrix). Stacking the columns of Y and replicating X won't do what you write; it forces each univariate regression to have the same coefficients.

Hi Folks, Thanks to several people's suggestions and clarifications, I think I have implemented a function which computes the conditional mean and covariance matrix of a subset of the dimensions of an MV-normal variable, given the values on the other dimensions (these conditioning value can be presented as a matrix, to deal with several cases at once). The code is below, for anyone who would

Hello, I have created a vector with 2 elements(see code below) I am calling this function many thousands of times (hundreds of thousands) after some time i get *** caught segfault *** address 0x5, cause 'memory not mapped' However, if i dont set the R_NamesSymbol, I do not get any such error. Am I doing this correctly? Thank you Saptarshi ==CODE=== // kxp and usar are two SEXP's

Here are another three problems with logarithmic scales: 1) segments() does not work with logarithmic scales. I suggest to change lines 962-973 in "plot.c": for (i = 0; i < n; i++) { if (FINITE(xt(x0[i%nx0])) && FINITE(yt(y0[i%ny0])) && FINITE(xt(x1[i%nx1])) && FINITE(yt(y1[i%ny1]))) { GP->col = INTEGER(col)[i % ncol];

Hello list, I am trying to test a model but for the beginning I want to do this by using simulated dataset. The model is Y_t = X_t %*% beta + e Where Y : (Nx1); X: (Nxp); beta: (0.6,0.3,0.1); e-uncorrelated normally distributed variates for each t. and later I want to use to use this dataset in a BUGS model to estimate the betas. Thank you for you consideration. [[alternative

Dear all, I have a simple question which I unfortunately do not seem to be able to solve myself. I have a (NxK) matrix and want to generate a new matrix by multiplying each row with itself such that the new matrix has dimension ((N*K)xK) (or better, generate an array with dimension (K,K,N)). I tried apply, but that did not work. Any suggestions? Thanks! Stephan ## Here is a simple

Dear all R users, I wanted to calculated a sample Variance covariance matrix of a five-variate normal distribution. However I stuck to calculate each element of that matrix. My question is should I calculate ordinary variance and covariances, taking pairwise variables? or I should take partial covariance between any two variables, keeping other fixed. In my decent opinion is I should go for the

I'm trying to make multiple line plots, each with a different color, using the xyplot command. Specifically, I have an NxK matrix Y and an Nx1 matrix x. I would like the plot to contain a line for each (x, Y[,i]), i=1:K. I know something like xyplot(Y[,1] + Y[,2] + Y[,3] ~ x, type='l') will work, but if Y is large, this notation can get very awkward. Is there a way to do something

Hi Folks, Given k RVs with MVN distribution N(mu,S) (S a kxk covariance matrix), let (w.l.o.g.) X1 denote the first r of them, and X2 the last (k-r). Likewise, let mu1 and mu2 denote their respective expectations. Then, of course, the expectation of X2 given X1=x1 is mu2 + S21*inv(S22)*(x1 - mu1) and the covariance matrix of X2 given X1=x2 is S22 - S21*inv(X11)*S12 where Sij is the

Hi I have the following problem: given x (px1) and S (pXp positive definite), find y such that y_i<=0 (i=1..p) minimizing the mahalanobis distance (x-y)'S^{-1}(x-y). Has anyone worked on this problem? Tips or R code would be appreciated. David ____________________ David Edwards Principal scientist Biostatistics Novo Nordisk A/S Novo Allé 2880 Bagsvaerd Denmark +45 4444 8888 (phone)

Hi, has anyone ever seen implemented in R the following "geodesic" distance between positive definite pxp matrices A and B? d(A,B) = \sum_{i=1}^p (\log \lambda_i)^2 were \lambda is the solution of det(A -\lambda B) = 0 thanks stefano

Thanks to Ross Ihaka and Bob Wheeler for responding to my earlier question. I looked into the Johnson system functions in SuppDists package. For now, I want to stick with the density(...) estimator, and so still need the variate lookup function. As per Ross' suggestion, I just did a numerical integration on the density object and used approxfun/splinefun to "lookup" the variate

> -----Original Message----- > From: r-help-bounces at stat.math.ethz.ch > [mailto:r-help-bounces at stat.math.ethz.ch] On Behalf Of > Ted.Harding at nessie.mcc.ac.uk > Sent: Sunday, December 05, 2004 7:14 PM > To: r-help at stat.math.ethz.ch > Subject: Re: [R] What is the most useful way to detect > nonlinearity in lo > > > On 05-Dec-04 Peter Dalgaard wrote:

Hi, I want to simulate multivariate data with > 1 distribution type. For example, I would like one normal variate and one poisson variate with a specified correlation structure. Is there a package that has this implemented? Thanks. Chuck [[alternative HTML version deleted]]

Hi Folks, Given a series of n independent Bernoulli trials with outcomes Yi (i=1...n) and Prob[Yi = 1] = Pi, I want P = Prob[sum(Yi) = r] (r = 0,1,...,n) I can certainly find a way to do it: Let p be the vector c(P1,P2,...,Pn). The cases r=0 and r=n are trivial (and also are exceptions for the following routine). For a given value of r in (1:(n-1)), library(combinat) Set <- (1:n)

Hi, I finally managed to make it work (I just needed to have a X11 window open). Thank you very much for your help. Agnes -----Original Message----- From: Ted Harding [mailto:Ted.Harding at nessie.mcc.ac.uk] Sent: Friday, July 21, 2006 3:34 PM To: Paquet, Agnes Cc: r-help at stat.math.ethz.ch Subject: RE: [R] connection to X11 problem On 21-Jul-06 Paquet, Agnes wrote: > Dear List, >

I'm following up on a question I posted 8/10/2010, but my newsreader has lost this thread. > Barrett & Ling, JASA, 1992, v.87(417), pp184-191 define general > classes of influence measures for multivariate > regression models, including analogs of Cook's D, Andrews & Pregibon > COVRATIO, etc. As in univariate > response models, these are based on leverage and

How about: nis.na <- complete.cases --- From: <Ted.Harding at nessie.mcc.ac.uk> Hi Folks, My recent response to Laura Quinn's query about matrix subsetting reminded of a question. I wrote: iDir <- ((Winds[,20]<45)|(Winds[,20]>315))&(!is.na(Winds[,20])) Now, I find "!is.na" a bit awkward to type, so I might prefer to type it as "nis.na".

Hi, Could anyone inform how to perform multi-variate auto regression using the past 't' values for regression in R. I have looked at ARMA provided by DES library and mvr provided by PLS library but could not match them to my requirements. Specifically, I want the following Say I have attributes a1-a4. and the regression equation is as follows: a4(t) =

Dear all, I am looking for routines which allow multi-variate sampling from non-normal distributions (loglogistic) given correlations among the variables. Unfortunately, I could not find a suitable package for R. Does anybody know one? Many thanks and best regards, Stefan Albrecht [[alternative HTML version deleted]]