conf.set {conf.design} | R Documentation |
Find minimal complete sets of confounded effects from a defining set for symmetric confounded factorial designs. Useful for checking if a low order interaction will be unintentionally confounded with blocks. As in the usual convention, only effects whose leading factor has an index of one are listed.
conf.set(G, p)
G |
Matrix whose rows define the effects to be confounded with blocks, in the same way as for conf.design(). |
p |
Number of levels for each factor. Must be a prime number. |
The function constructs all linear functions of the rows of G (over GF(p)), and removes those rows whose leading non-zero component is not one.
A matrix like G with a minimal set of confounded with blocks defined in the rows.
None
conf.design
G <- rbind(c(1,2,1,0), c(0,1,1,1)) dimnames(G) <- list(NULL, LETTERS[1:4]) conf.set(G, 3) # A B C D # [1,] 1 2 1 0 # [2,] 0 1 1 1 # [3,] 1 0 2 1 # [4,] 1 1 0 2 # If A B^2 C and B C D are confounded with blocks, then so are A C^2 D # and A B D^2. Only three-factor interactions are confounded, so the # design is presumably useful.