A coding for factors based on successive differences.
contr.sdif(n, contrasts = TRUE, sparse = FALSE)
If contrasts
is TRUE
, a matrix with n
rows and
n - 1
columns, and the n
by n
identity matrix if
contrasts
is FALSE
.
The number of levels required.
logical: Should there be n - 1
columns orthogonal to the mean
(the default) or n
columns spanning the space?
logical. If true and the result would be sparse (only
true for contrasts = FALSE
), return a sparse matrix.
The contrast coefficients are chosen so that the coded coefficients in a one-way layout are the differences between the means of the second and first levels, the third and second levels, and so on. This makes most sense for ordered factors, but does not assume that the levels are equally spaced.
Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics with S. Fourth Edition, Springer.
(A <- contr.sdif(6))
zapsmall(ginv(A))
Run the code above in your browser using DataLab