Simple Bayesian linear model with non-informative priors

Given a lm object, the bayesLMRef function fits a simple Bayesian linear model with reference (non-informative) priors.

Usage

bayesLMRef(lm.obj, n.samples, ...)

Arguments

lm.obj

an object returned by lm.

n.samples

the number of posterior samples to collect.

...

currently no additional arguments.

Value

An object of class bayesLMRef, which is a list with at least the following tag:

p.beta.tauSq.samples

a coda object of posterior samples for the defined parameters.

Details

See page 355 in Gelman et al. (2004).

References

Gelman, A., Carlin, J.B., Stern, H.S., and Rubin, D.B. (2004). Bayesian Data Analysis. 2nd ed. Boca Raton, FL: Chapman and Hall/CRC Press.

Author

Sudipto Banerjee sudiptob@biostat.umn.edu,
Andrew O. Finley finleya@msu.edu

Examples

if (FALSE) { # \dontrun{
set.seed(1)

n <- 100
X <- as.matrix(cbind(1, rnorm(n)))
B <- as.matrix(c(1,5))
tau.sq <- 0.1
y <- rnorm(n, X%*%B, sqrt(tau.sq))

lm.obj <- lm(y ~ X-1)

summary(lm.obj)

##Now with bayesLMRef
n.samples <- 500

m.1 <- bayesLMRef(lm.obj, n.samples)

summary(m.1$p.beta.tauSq.samples)
} # }