R help - Nov 2008 - Generating Data for Simulation

Are there any R packages that could be used to generate random data given a
set of parameters?  Or, if not a package, how would one generate such data?
What I would like to do is simulate some sample data for a regression model
given a set of population covariances and distribution parameters to be used
in a MC simulation.

Thanks for any tips.

	[[alternative HTML version deleted]]

Brett Magill <magillb <at> sbcglobal.net> writes:
> 
> Are there any R packages that could be used to generate random data given a
> set of parameters?  Or, if not a package, how would one generate such data?
> What I would like to do is simulate some sample data for a regression model
> given a set of population covariances and distribution parameters to be
used
> in a MC simulation.
> 
> Thanks for any tips.
> 
  This is very definitely doable, don't know of any end-user
packages to do it.

  Simple example: generate data according to

n=100
X ~ Uniform(0,10)
Y ~ Normal(1+2*X,sigma=0.1)

set.seed(1001) ## for reproducibility
n <- 100
x <- runif(n,0,10)
y <- rnorm(n,mean=1+2*x,sigma=0.1)

  You can look at chapter 5 of
http://www.zoology.ufl.edu/bolker/emdbook/book.pdf
(or buy the book) for more information ...

  cheers
    Ben Bolker