normal.inverse.wishart.prior: Normal inverse Wishart prior

In Boom: Bayesian Object Oriented Modeling

Normal inverse Wishart prior

Description

The NormalInverseWishartPrior is the conjugate prior for the mean and variance of the multivariate normal distribution. The model says that

Σ^{-1} ~ Wishart(ν, S) μ | Σ ~ N(μ0, Σ/κ)

The Wishart(S, ν) distribution is parameterized by S, the inverse of the sum of squares matrix, and the scalar degrees of freedom parameter nu.

The distribution is improper if ν < dim(S).

Usage

NormalInverseWishartPrior(mean.guess,
                          mean.guess.weight = .01,
                          variance.guess,
                          variance.guess.weight = nrow(variance.guess) + 1)

Arguments

mean.guess	The mean of the prior distribution. This is μ0 in the description above.
mean.guess.weight	The number of observations worth of weight assigned to mean.guess. This is κ in the description above.
variance.guess	A prior estimate at the value of Σ. This is S^{-1}/ν in the notation above.
variance.guess.weight	The number of observations worth of weight assigned to variance.guess. This is df.

Author(s)

Steven L. Scott [e-mail]

References

Gelman, Carlin, Stern, Rubin (2003), "Bayesian Data Analysis", Chapman and Hall.

Boom documentation built on Nov. 10, 2022, 5:56 p.m.