cprob: Extract Contour Probabilities

Description

Function to extract estimated contour probabilities of a particular effect estimated with P-splines using Markov chain Monte Carlo (MCMC) estimation techniques. Note that, the contour probability option must be specified within function sx, see the example.

Usage

cprob(object, model = NULL, term = NULL, ...)

Arguments

object: an object of class "bayesx".
model: for which model the contour probabilities should be provided, either an integer or a character, e.g. model = "mcmc.model".
term: if not NULL, the function will search for the term contour probabilities should be extracted for, either an integer or a character, eg term = "s(x)".
...: not used.

Author

Nikolaus Umlauf, Thomas Kneib, Stefan Lang, Achim Zeileis.

References

Brezger, A., Lang, S. (2008): Simultaneous probability statements for Bayesian P-splines. Statistical Modeling, 8, 141--186.

Examples

Run this code

if (FALSE) {
## generate some data
set.seed(111)
n <- 500

## regressor
dat <- data.frame(x = runif(n, -3, 3))

## response 
dat$y <- with(dat, 1.5 + sin(x) + rnorm(n, sd = 0.6))

## estimate model
## need to set the contourprob option, 
## otherwise BayesX will not calculate probabilities
## see also the reference manual of BayesX available
## at www.BayesX.org
b <- bayesx(y ~ sx(x, bs = "ps", contourprob = 4), data = dat)

## extract contour probabilities
cprob(b, term = "sx(x)")
}