rvmultinom {rv}R Documentation

Generate Random Variables from a Multinomial Sampling Model

Description

Generates a random vector from a multinomial sampling model.

Usage

  rvmultinom(n=1, size=1, prob)

Arguments

n integer, number of random variables to generate
size integer or integer-valued rv: the number of trials (size of each sample)
prob vector (of length at least 3) prior probabilities of successes of each trial (may be constant or an rv object)

Details

The length of prob determines the number of bins.

The vector prob will be normalized to have sum 1.

If length(prob) is two, rvbinom is called instead.

NOTE. Case of random n or size or prob — not yet optimized for speed.

Value

A random array of dimensions length(prob) times n.

Author(s)

Jouni Kerman kerman@stat.columbia.edu http://www.stat.columbia.edu/~kerman

References

Kerman, Jouni and Gelman, Andrew. Manipulating and Summarizing Posterior Simulations Using Random Variable Objects. Technical report, Columbia University, New York.

Examples

  y <- rvmultinom(n=3, size=1, prob=c(0.20, 0.30, 0.50))

[Package rv version 0.949 Index]