contr.full: Contrasts where each factor-level is explicitly represented

In jwdink/survbayes: Tools for bayesian survival-regression, with Stan and MAP

Description

This contrast-type should only be used in survreg_map, where (a) the function knows to center the contrasts so that each coefficient estimate can be interpreted as "deviation from average," and (b) the fact that this contrast-type is linearly dependent isn't a problem, thanks to shrinkage from priors.

Usage

contr.full(n, contrasts = TRUE, sparse = FALSE, f = NULL)

Arguments

n	a vector of levels for a factor, or the number of levels.
contrasts	Just set this to TRUE
sparse	logical indicating if the result should be sparse
f	The actual factor vector

Value

Contrast-matrix

jwdink/survbayes documentation built on May 4, 2019, 4:23 p.m.