DirichSampHWE: Simulate samples from a Dirichlet prior or posterior under...

In HWEBayes: Bayesian investigation of Hardy-Weinberg Equilibrium via estimation and testing.

Description

Function to simulate samples from the HWE Dirichlet model. Can be used for samples from the prior or the (conjugate) Dirichlet posterior, both in the k allele case. Samples are generated for the allele frequencies in the order p_{1},p_{2},...,p_{k}.

Usage

DirichSampHWE(nvec, bvec0, nsim)

Arguments

nvec	vector of genotype frequencies in the order n_{11}, n_{12},..., n_{1k},n_{22} ..., n_{2k},..., n_{kk}.
bvec0	vector of length k Dirichlet prior parameters, where k is the number of alleles.
nsim	number of samples to simulate from the prior/posterior.

Details

Uses the rdirichlet function from the MCMCpack library.

Value

pvec	matrix of size nsim \times k containing samples for the genotype frequencies, in the order p_{1}, p_{12},..., p_{k}.

Author(s)

Jon Wakefield ([e-mail]).

References

Wakefield, J. (2010). Bayesian methods for examining Hardy-Weinberg equilibrium. Biometrics; Vol 66:257-65

See Also

DirichSampSat, DirichNormSat, DirichNormHWE

Examples

# First sample from the prior
PriorSampHWE <- DirichSampHWE(nvec=rep(0,10),bvec0=rep(1,4),nsim=1000)
par(mfrow=c(1,1))
hist(PriorSampHWE$pvec[,1],xlab="p1",main="")
# Now sample from the posterior
data(DiabRecess)
PostSampHWE <- DirichSampHWE(nvec=DiabRecess,bvec0=rep(1,4),nsim=1000)
par(mfrow=c(1,1))
hist(PostSampHWE$pvec[,1],xlab="p1",main="")

HWEBayes documentation built on May 2, 2019, 11:07 a.m.