R help - Feb 2007 - Generating MVN Data

Dear All

I want to generate multivariate normal data in R for a given covariance 
matrix, i.e. my generated data must have the given covariance matrix. I 
know the rmvnorm command is to be used but may be I am failing to 
properly assign the covariance matrix.

Any help will be greatly appreciated

thanks.

M. R. Ahmad

Rauf Ahmad wrote:
> Dear All
> 
> I want to generate multivariate normal data in R for a given covariance 
> matrix, i.e. my generated data must have the given covariance matrix. I 
> know the rmvnorm command is to be used but may be I am failing to 
> properly assign the covariance matrix.
> 
> Any help will be greatly appreciated
library(MASS)

mat <- mvrnorm(100, mu=rep(0,4), Sigma = diag(4), empirical=TRUE)

cov(mat)
              [,1]          [,2]          [,3]          [,4]
[1,]  1.000000e+00 -1.634448e-16 -1.004223e-16 -2.015521e-16
[2,] -1.634448e-16  1.000000e+00  4.244391e-17  1.544399e-17
[3,] -1.004223e-16  4.244391e-17  1.000000e+00 -1.921951e-16
[4,] -2.015521e-16  1.544399e-17 -1.921951e-16  1.000000e+00

?mvrnorm
> thanks.
> 
> M. R. Ahmad
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
-- 
Chuck Cleland, Ph.D.
NDRI, Inc.
71 West 23rd Street, 8th floor
New York, NY 10010
tel: (212) 845-4495 (Tu, Th)
tel: (732) 512-0171 (M, W, F)
fax: (917) 438-0894

Hi

give rmvnorm() any symmetric positive definite matrix and it should  
work:



 > rmvnorm(n=10,mean=1:2,sigma=matrix(c(1,0.5,0.5,1),2,2))
          [,1]   [,2]
[1,] -0.1118  2.514
[2,]  1.8667  1.628
[3,]  3.2477  2.263
[4,]  1.0166  2.381
[5,] -0.0888 -0.132
[6,] -0.9249  0.610
[7,]  1.5046  3.578
[8,]  0.8530  0.802
[9,]  2.2940  2.240
[10,]  1.1660  2.528
 >



HTH

rksh


On 13 Feb 2007, at 14:14, Rauf Ahmad wrote:
> Dear All
>
> I want to generate multivariate normal data in R for a given  
> covariance
> matrix, i.e. my generated data must have the given covariance  
> matrix. I
> know the rmvnorm command is to be used but may be I am failing to
> properly assign the covariance matrix.
>
> Any help will be greatly appreciated
>
> thanks.
>
> M. R. Ahmad
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting- 
> guide.html
> and provide commented, minimal, self-contained, reproducible code.
--
Robin Hankin
Uncertainty Analyst
National Oceanography Centre, Southampton
European Way, Southampton SO14 3ZH, UK
  tel  023-8059-7743

you probably want to use mvrnorm() from package MASS, e.g.,

library(MASS)
mu <- c(-3, 0, 3)
Sigma <- rbind(c(5,3,2), c(3,4,1), c(2,1,3))
x <- mvrnorm(1000, mu, Sigma, empirical = TRUE)

colMeans(x)
var(x)


I hope it helps.

Best,
Dimitris

----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://med.kuleuven.be/biostat/
     http://www.student.kuleuven.be/~m0390867/dimitris.htm


----- Original Message ----- 
From: "Rauf Ahmad" <rahmad at gwdg.de>
To: <r-help at stat.math.ethz.ch>
Sent: Tuesday, February 13, 2007 3:14 PM
Subject: [R] Generating MVN Data

> Dear All
>
> I want to generate multivariate normal data in R for a given 
> covariance
> matrix, i.e. my generated data must have the given covariance 
> matrix. I
> know the rmvnorm command is to be used but may be I am failing to
> properly assign the covariance matrix.
>
> Any help will be greatly appreciated
>
> thanks.
>
> M. R. Ahmad
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide 
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
> 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm

Dear All

many thanks for your kind help and quick response. I actually used the 
same code (but as rmvnorm in library mvtnorm) to generate data. But I 
was not sure since I thought I might have to do some decomposition of 
the matrix and then generate the data. But now it is confirmed.

with many thanks again and regards

M. R. Ahmad


Dear All

I want to generate multivariate normal data in R for a given covariance
matrix, i.e. my generated data must have the given covariance matrix. I
know the rmvnorm command is to be used but may be I am failing to
properly assign the covariance matrix.

Any help will be greatly appreciated

thanks.

M. R. Ahmad