arm (version 1.15-1)

model.matrixBayes: Construct Design Matrices

Description

model.matrixBayes creates a design matrix.

Usage

model.matrixBayes(object, data = environment(object),
    contrasts.arg = NULL, xlev = NULL, keep.order = FALSE, drop.baseline=FALSE,...)

Details

model.matrixBayes is adapted from model.matrix in the stats pacakge and is designed for the use of bayesglm. It is designed to keep baseline levels of all categorical varaibles and keep the variable names unodered in the output. The design matrices created by model.matrixBayes are unidentifiable using classical regression methods, though; they can be identified using bayesglm.

References

Andrew Gelman, Aleks Jakulin, Maria Grazia Pittau and Yu-Sung Su. (2009). “A Weakly Informative Default Prior Distribution For Logistic And Other Regression Models.” The Annals of Applied Statistics 2 (4): 1360--1383. https://sites.stat.columbia.edu/gelman/research/published/priors11.pdf

See Also

model.frame, model.extract, terms, terms.formula, bayesglm.

Examples

ff <- log(Volume) ~ log(Height) + log(Girth)
str(m <- model.frame(ff, trees))
(model.matrix(ff, m))
class(ff) <- c("bayesglm", "terms", "formula")
(model.matrixBayes(ff, m))