rdirichlet {bayess} | R Documentation |
Random generator for the Dirichlet distribution
Description
This function simulates a sample from
a Dirichlet distribution on the k
dimensional simplex with
arbitrary parameters.
The simulation is based on a renormalised vector of gamma variates.
Usage
rdirichlet(n = 1, par = rep(1, 2))
Arguments
n |
number of simulations |
par |
parameters of the Dirichlet distribution, whose length determines
the value of |
Details
Surprisingly, there is no default Dirichlet distribution generator in the R base
packages like MASS
or stats
. This function can be used in full
generality, apart from the book (Chapter 6).
Value
returns a (n,k)
matrix of Dirichlet simulations
Examples
apply(rdirichlet(10,rep(3,5)),2,mean)
[Package bayess version 1.6 Index]