search for: multicarri

I sent this to R-devel early last month, but have received no response, so I guess it really is a bug. This looks like a bug to me, and is a bit hard to describe, but easy to reproduce. ? Basically, if RNGkind is saved as something other than the default, and if the first operation in a session is a set.seed(), the default is reverted to. ?Reproduce by: cafe-rozo> ?R --vanilla R version

Dear all, Recently I am generating large random samples (10M) and any duplicated numbers are not desired. We tried several RNGs in R and found Wichmann-Hill did not produce duplications. The duplication problem is the interesting birthday problem. If there are M possible numbers, randomly draw N numbers from them, the average number of dupilcations D = N(N-1)/2/M. For Knuth-TAOCP and

This looks like a bug to me, and is a bit hard to describe, but easy to reproduce. Basically, if RNGkind is saved as something other than the default, and if the first operation in a session is a set.seed(), the default is reverted to. Reproduce by: cafe-rozo> R --vanilla R version 2.7.1 (2008-06-23) Copyright (C) 2008 The R Foundation for Statistical Computing ISBN 3-900051-07-0 R is

I am writing a function for the purposes of a simulation. Due to memory problems, the function sometimes crashes. In order to get around this problem, I would like to include to be able to save the "last" seed, so I can pick up with the next run of the simulation after a "crash". I am having trouble understanding what is going on with .Random.seed! For each run of the

Dear Sir/Madam, I am a beginner in R. Here is my questions. 1. Can you give me one test for randomness (a name and descriptive paragraph is sufficient). 2. I have learned a uniform random number generator [e.g. not the algorithms: i)Wichmann-Hill, ii) Marsaglia-Multicarry, iii) Super-Duper (Marsaglia), iv) Mersenne-Twister, v) TAOCP-1997 (Knuth), or vi) TAOCP-2002 (Knuth)] . Is there any other

I have two questions about using a user-defined random number generator (RNG) with the standalone Rmath library. The default RNG with the standalone Rmath library is the Marsaglia-multicarry generator, which has poor properties. The "R Installation and Administration" manual, in the section "The standalone Rmath library", states that: ``` A little care is needed to use the

Hi, I have a few questions about the RNG in R; apologies if these are dumb questions: 1. It is my understanding that, among the three types of random number generators available in R now, the best one is the Marsaglia Multicarry. Is this correct? 2. How does the best RNG in R compare (in terms of quality) to the RNG in SPlus? (based on Marsaglia's Super Duper)? Does the Super-Duper in

R 0.64 on windows NT 4.0 Sometimes I got an error message by doing this > .Random.seed <- c(1, 1:2) > .Random.seed [1] 1 1 2 > runif(5) Warning: Wrong length .Random.seed; forgot initial RNGkind? set to Wichmann-Hill[1] 0.02253721 0.84832584 ........ Sometimes I do not get error message: > .Random.seed <- c(1, 1:2) > .Random.seed [1] 1 1 2 > runif(1) [1] 0.5641106 >

I'm running into problems with set.seed--maybe I'm misunderstanding something. I'm running R 1.5.1 on Windows 2000. I'm basically trying to capture the random seed so that I can reproduce a simulation if it's necessary later. Using set.seed, I can certainly get reproducible results, but not the results I get on the first pass. Here's an example: # Generate a random

Hi R-Developers, I had a question about the random number generator used in the R StandAlone Math Library. The stand-alone library depends on the unif_rand() function for most simulated values, and this function is provided in the sunif.c file in the relevant directory. At present, this program implements the "Marsaglia-Multicarry" algorithm, which is described throughout the R

Hi, I'm seeing something that may be a bug in R's standalone math library, which is packaged by Debian as r-mathlib. I reported it to the Debian BTS as http://bugs.debian.org/657573 I'm using Debian squeeze, and the code was tested with r-mathlib 2.11.1-6 (default on stable) and 2.14.1-1 (from testing/unstable). I summarize this report below. The following code with the R math

This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. ---559023410-851401618-1218751024=:15885 Content-Type: TEXT/PLAIN; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE I didn't describe the problem clearly. It's about the number of distinct=20 values. So just

Hello everyone. If I do f <- function(n){max(rnorm(n))} plot(sapply(rep(5000,4000),f)) #[this takes my PC about 30 seconds] then I get something quite unexpected: gaps in the distribution. For me, the most noticable one is at about 3.6. Do others get this? Is it an optical illusion? It can't be right, can it? Or maybe I just don't understand the good ol' Gaussian very

Dear All, I want to use R's random number in a C++ program (I can link libRmath either as shared or static library). I have two questions: 1. If I understand correctly, the underlaying random number generator will be Marsaglia-multicarry, UNLESS I provide my own. In other words (unless I provide it) I cannot use some of the other RNG's available from within R, such as Mersenne-Twister?

Recieving digests. > RNGkind(NULL) [1] "Marsaglia-Multicarry" "Kinderman-Ramage" I would appreciate it if anybody has any comments on the following. Please do not comment on the R functions themselves, since they merely mimic a (bivariate simplification of a) C routine called from S. In particular, I would like to know if anything is available with regard to the

I have applied L'Ecuyer's TESTU01 suite of RNG tests to the RNGs in R. TESTU01 offers three increasingly more stringent suites, called "Small Crush", "Crush" and "Big Crush". If a particular RNG fails Small Crush, there is no need to apply Big Crush. Below I summarize the results: Number of Tests Failed Small Crush Crush Big Crush

I have commited fixes for 0.64.1 for (1) From: Mai Zhou <mai@ms.uky.edu> > .Random.seed <- c(1, 1:2) > .Random.seed [1] 1 1 2 > runif(5) Warning: Wrong length .Random.seed; forgot initial RNGkind? set to Wichmann-Hill[1] 0.02253721 0.84832584 ........ Here the length of the seed was being tested before the kind was picked out, so the length of the previous type was used. (2)

Brian I've been playing a bit with the Wichmann-Hill RNG. I would prefer to have normally distributed random numbers and I think I have things generally worked out to use Wichmann-Hill and then Box-Muller. In the process, I was looking at R's rnorm.c, but could not figure out what transformation is used in R to convert uniform rv's to normal rv's. Do you know? It looks like there

This question may not be specific to R, but I'm using R so here goes: Since R is slow (as is Splus) I want to split a simulation and run it on 2 or 3 systems at once. The simulations involve generating a large number of random values. How can I set .Random.seed so that the succession of random values don't overlap across systems. I see that when I invoke R and give command runif(1) a

Dear People, I have a vexing problem related to pseudo-random number generation, and would appreciate any help and advice. This problem is not directly related to R, and the only reason I am posting it to this list is that my implementation is using R. Let me describe my problem by giving an example, that is close to what I am trying to do. Suppose we are given a stream of pseudo-random numbers,