contrasts

0th

Percentile

Get and Set Contrast Matrices

Set and view the contrasts associated with a factor.

Keywords
regression, design
Usage
contrasts(x, contrasts = TRUE, sparse = FALSE)
contrasts(x, how.many) <- value
Arguments
x
a factor or a logical variable.
contrasts
logical. See ‘Details’.
sparse
logical indicating if the result should be sparse (of class dgCMatrix), using package Matrix">https://CRAN.R-project.org/package=Matrix.
how.many
How many contrasts should be made. Defaults to one less than the number of levels of x. This need not be the same as the number of columns of value.
value
either a numeric matrix (or a sparse or dense matrix of a class extending dMatrix from package Matrix">https://CRAN.R-project.org/package=Matrix) whose columns give coefficients for contrasts in the levels of x, or the (quoted) name of a function which computes such matrices.
Details

If contrasts are not set for a factor the default functions from options("contrasts") are used. A logical vector x is converted into a two-level factor with levels c(FALSE, TRUE) (regardless of which levels occur in the variable). The argument contrasts is ignored if x has a matrix contrasts attribute set. Otherwise if contrasts = TRUE it is passed to a contrasts function such as contr.treatment and if contrasts = FALSE an identity matrix is returned. Suitable functions have a first argument which is the character vector of levels, a named argument contrasts (always called with contrasts = TRUE) and optionally a logical argument sparse. If value supplies more than how.many contrasts, the first how.many are used. If too few are supplied, a suitable contrast matrix is created by extending value after ensuring its columns are contrasts (orthogonal to the constant term) and not collinear.

References

Chambers, J. M. and Hastie, T. J. (1992) Statistical models. Chapter 2 of Statistical Models in S eds J. M. Chambers and T. J. Hastie, Wadsworth & Brooks/Cole.

C, contr.helmert, contr.poly, contr.sum, contr.treatment; glm, aov, lm.
library(stats) utils::example(factor) fff <- ff[, drop = TRUE] # reduce to 5 levels. contrasts(fff) # treatment contrasts by default contrasts(C(fff, sum)) contrasts(fff, contrasts = FALSE) # the 5x5 identity matrix contrasts(fff) <- contr.sum(5); contrasts(fff) # set sum contrasts contrasts(fff, 2) <- contr.sum(5); contrasts(fff) # set 2 contrasts # supply 2 contrasts, compute 2 more to make full set of 4. contrasts(fff) <- contr.sum(5)[, 1:2]; contrasts(fff) ## using sparse contrasts: % useful, once model.matrix() works with these : ffs <- fff contrasts(ffs) <- contr.sum(5, sparse = TRUE)[, 1:2]; contrasts(ffs) stopifnot(all.equal(ffs, fff)) contrasts(ffs) <- contr.sum(5, sparse = TRUE); contrasts(ffs)