similar to: Gelman-Rubin Convergence test

I have been running a Gibbs Sampler to estimate levels of efficiency in the Louisiana Shrimp Industry. I created a matrix (samp) where I stored the results of each iteration for 86 variables. I run 10,000 iterations. So, the matrix samp is 10,000 x 86. I want to use the gelman-rubin test to check for convergence. To do that, I need at least two chains. If I run second chain with different starting

I was trying to generate random numbers with a gamma distribution. In R the function is: rgamma(n, shape, rate = 1, scale = 1/rate). My question is that if X~gamma(alpha, beta) and I want to generate one random number where do I plug alpha and beta in rgamma? and, what is the meaning and use of rate? Thanks for your attention, Jorge [[alternative HTML version deleted]]

Hi, A question. When I run gelman.diag and gelman.plot with mcmc lists obtained from MCMCregress, the results are following. > post.R <- MCMCregress(Size~Age+Status, data = data, burnin = 5000, mcmc = 100000, + thin = 10, verbose = FALSE, beta.start = NA, sigma2.start = NA, + b0 = 0, B0 = 0, nu = 0.001, delta = 0.001) > post1.R <- MCMCregress(Size~Age+Status, data

Dear, I'm trying to run diagnostics on MCMC analysis (fitting a log-linear model to rates data). I'm getting an error message when trying Gelman-Rubin shrink factor plot: >gelman.plot(out) Error in chol.default(W) : the leading minor of order 2 is not positive definite I take it that somewhere, somehow a matrix is singular, but how can that be remedied? My code: library(rjags)

Dear friends, Over the past weeks, I have been asking a lot of questions about how to use R in Bayesian analysis. I am brand new to R, but I am very pleased with it. I started with winbugs but I found winbugs to be a limited software, not bad but has several limitations. By contrast, R allows the analyst to tackle any problem with a huge set of tools for any kind of analysis. I love R. In

Dear Friends, I am trying to use the gelman-rubin convergence test. I generated a matrix samp[10,000x86] with the gibbs sampler. the test requires the creation of "mcmc" objects. Since I don't know how to define samp as a "mcmc" object, I tried to create one mcmc object by means of the mcmc() function. With this function I tried to create a mcmc object dul from samp but I

I am using -mice- for multiple imputation and would like to use the gelman diagnostic in -coda- to assess the convergence of my imputations. However, gelman.diag requires an mcmc list as input. van Buuren and Groothuis-Oudshoorn (2011) recommend running mice step-by-step to assess convergence (e.g. imp2 <- mice.mids(imp1, maxit = 3, print = FALSE) ) but this creates mids objects. How can I

Hi, I am trying to recreate the right graph on page 524 of Gelman's 2006 paper "Prior distributions for variance parameters in hierarchical models" in Bayesian Analysis, 3, 515-533. I am only interested, however, in recreating the portion of the graph for the overlain prior density for the half-Cauchy with scale 25 and not the posterior distribution. However, when I try:

Deal all, as MCMClogit does not allow for the specification of several chains, I have run my model 3 times with different random number seeds and differently dispersed multivariate normal priors. For example: res1 = MCMClogit(y~x,b0=0,B0=0.001,data=mydat, burnin=500, mcmc=5500, seed=1234, thin=5) res2 = MCMClogit(y~x,b0=1,B0=0.01,data=mydat, burnin=500, mcmc=5500, seed=5678, thin=5) res3 =

i am having problem running my own data. yesterday it was working just fine. today it is not. this is the code i was using as an example to follow. this code ALSO worked just fine yesterday, and is no longer working at all. i suspect it is a problem with either my computer or the software, at this point. if THIS won't even run.... something is wrong. i can assure you this isn't

Hello, I know of some various methods out there to utilize multiple processors but am not sure what the best solution would be. First some things to note: I'm running dependent simulations, so direct parallel coding is out (multicore, doSnow, etc). I'm on Windows, and don't know C. I don't plan on learning C or any of the *nix languages. My main concern deals with Multiple

Hi all, following is my R -code and shows the error given below > n <- 100 > k<-2 > x1 <-c(1, 3);x2 <- c(2,5) > X <- matrix(c(0,0), nrow = 2, ncol = n) > for(i in 1:k) + X[i, ] <- mh1.epidemic(n,x1[i],x2[i]) Error in X[i, ] <- mh1.epidemic(n,x1[i],x2[i]):   number of items to replace is not a multiple of replacement length   > psi <-t(apply(X, c(1,2),

Readers, Openjdk and ibm java versions have failed to install, all reporting a bad elf, e.g. ./ibm-java-i386-sdk-7.0-0.0.bin Preparing to install... Extracting the JRE from the installer archive... Unpacking the JRE... Extracting the installation resources from the installer archive... Configuring the installer for this system's environment... strings: '/lib/libc.so.6': No such file

OK, I spammed the mailing list recently, but I will be fired if I can''t solve the problem today. (just kidding, but I did waste lots of time on it :( ) The common configuration for teql is for two computers connected directly with two links. My topology is a little different: one link is connected directly, but the other is connected through a gateway. My problem is teql can''t

I'm just starting to learn to use R, and although I'm seeing lots of functions aimed at doing orthodox statistical analyses, I don't see the same for Bayesian analyses. What support does R have for Bayesian statistics?

Package: mcgibbsit Title: Warnes and Raftery's MCGibbsit MCMC diagnostic Version: 1.0 Author: Gregory R. Warnes <gregory_r_warnes at groton.pfizer.com> Description: mcgibbsit provides an implementation of Warnes & Raftery's MCGibbsit run-length diagnostic for a set of (not-necessarily independent) MCMC sampers. It combines the estimate error-bounding approach of Raftery

Package: mcgibbsit Title: Warnes and Raftery's MCGibbsit MCMC diagnostic Version: 1.0 Author: Gregory R. Warnes <gregory_r_warnes at groton.pfizer.com> Description: mcgibbsit provides an implementation of Warnes & Raftery's MCGibbsit run-length diagnostic for a set of (not-necessarily independent) MCMC sampers. It combines the estimate error-bounding approach of Raftery

R users- In moving from R 2.0.0 to R 2.1.0 in Windows, I have encountered a problem with the "riwish" command in the package "MCMCpack". I've searched the documentation and can't seem to figure it out. For example: Define a matrix: > lam <- matrix(c(.00233,-.00057,-.00057,.00190),2,2) and then use the riwish command to generate a random inverse-Wishart

Dear friends. On the Bayesian averaging homepage http://www.research.att.com/~volinsky/bma.html I found some S code some of which perhaps may run in R. There was a call to an algorithm possibly within S but not supported by R 64.1: "leaps and bounds". I guess it is a minimization step. Can anyone clarify the algorithm and perhaps even give a pointer to some code ? I guess this may be

Hello all, if I have multiple imputed data sets, is there a command or function in R in any package you know of to combine those, I know one common MI approach is rubins rules, is there a way to do this using his rules or others? I know theres ways, like using Amelia from Gary King's website to create the imputed data sets, but how to make them into one or combine them for analysis.