Re: [R] alternative generator for normal distributed variables

2012-05-30 Thread Jorge I Velez
Hi there, I am not an expert, but simple experimentation using the acf() function shows that your statement is not true (see ?acf and ?replicate for more information): foo <- function(x) acf(x, plot = FALSE)$acf[,,1] result <- replicate(1, foo(rnorm(100))) result <- t(apply(result, 1, quantil

Re: [R] alternative generator for normal distributed variables

2012-05-30 Thread R. Michael Weylandt
Like Thierrey I entirely disagree with you, but there are many PRNGs in R -- take a look at ?RNGkind to change them. Michael On Wed, May 30, 2012 at 7:39 AM, juliane0212 wrote: > Hello, > > currently I'm working on a model based on Monte-Carlo-Simulations. > > I observed that a generated normal

Re: [R] alternative generator for normal distributed variables

2012-05-30 Thread ONKELINX, Thierry
Please give a reproducible example when making bold statements. I find no evidence of autocorrelation in set.seed(12345) x <- rnorm(100, mean = 0, sd = 1) acf(x) x <- rnorm(1e6, mean = 0, sd = 1) acf(x) ir. Thierry Onkelinx Instituut voor natuur- en bosonderzoek / Research Institute for Nature