1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
|
\name{runiregGibbs}
\alias{runiregGibbs}
\concept{bayes}
\concept{Gibbs Sampler}
\concept{regression}
\concept{MCMC}
\title{Gibbs Sampler for Univariate Regression}
\description{
\code{runiregGibbs} implements a Gibbs Sampler to draw from posterior of a univariate regression with a conditionally conjugate prior.
}
\usage{runiregGibbs(Data, Prior, Mcmc)}
\arguments{
\item{Data }{list(y, X)}
\item{Prior}{list(betabar, A, nu, ssq)}
\item{Mcmc }{list(sigmasq, R, keep, nprint)}
}
\details{
\subsection{Model and Priors}{
\eqn{y = X\beta + e} with \eqn{e} \eqn{\sim}{~} \eqn{N(0, \sigma^2)}
\eqn{\beta} \eqn{\sim}{~} \eqn{N(betabar, A^{-1})}\cr
\eqn{\sigma^2} \eqn{\sim}{~} \eqn{(nu*ssq)/\chi^2_{nu}}
}
\subsection{Argument Details}{
\emph{\code{Data = list(y, X)}}
\tabular{ll}{
\code{y: } \tab \eqn{n x 1} vector of observations \cr
\code{X: } \tab \eqn{n x k} design matrix
}
\emph{\code{Prior = list(betabar, A, nu, ssq)} [optional]}
\tabular{ll}{
\code{betabar: } \tab \eqn{k x 1} prior mean (def: 0) \cr
\code{A: } \tab \eqn{k x k} prior precision matrix (def: 0.01*I) \cr
\code{nu: } \tab d.f. parameter for Inverted Chi-square prior (def: 3) \cr
\code{ssq: } \tab scale parameter for Inverted Chi-square prior (def: \code{var(y)})
}
\emph{\code{Mcmc = list(sigmasq, R, keep, nprint)} [only \code{R} required]}
\tabular{ll}{
\code{sigmasq: } \tab value for \eqn{\sigma^2} for first Gibbs sampler draw of \eqn{\beta}|\eqn{\sigma^2} \cr
\code{R: } \tab number of MCMC draws \cr
\code{keep: } \tab MCMC thinning parameter -- keep every \code{keep}th draw (def: 1) \cr
\code{nprint: } \tab print the estimated time remaining for every \code{nprint}'th draw (def: 100, set to 0 for no print)
}
}
}
\value{
A list containing:
\item{betadraw }{ \eqn{R x k} matrix of betadraws }
\item{sigmasqdraw }{ \eqn{R x 1} vector of sigma-sq draws}
}
\author{Peter Rossi, Anderson School, UCLA, \email{perossichi@gmail.com}.}
\references{For further discussion, see Chapter 3, \emph{Bayesian Statistics and Marketing} by Rossi, Allenby, and McCulloch.}
\seealso{ \code{\link{runireg}} }
\examples{
if(nchar(Sys.getenv("LONG_TEST")) != 0) {R=1000} else {R=10}
set.seed(66)
n = 200
X = cbind(rep(1,n), runif(n))
beta = c(1,2)
sigsq = 0.25
y = X\%*\%beta + rnorm(n,sd=sqrt(sigsq))
out = runiregGibbs(Data=list(y=y, X=X), Mcmc=list(R=R))
cat("Summary of beta and Sigmasq draws", fill=TRUE)
summary(out$betadraw, tvalues=beta)
summary(out$sigmasqdraw, tvalues=sigsq)
## plotting examples
if(0){plot(out$betadraw)}
}
\keyword{regression}
|
