planor (version 1.5-3)

alias-methods: Summarize the Design Properties

Description

Methods to summarize the design properties of an object containing key matrices. Display the design keys matrix(ces) and the factorial effects confounded with the mean.

Usage

# S4 method for designkey
alias(object, model, …)
# S4 method for keymatrix
alias(object, model, fact, block, …)
# S4 method for listofdesignkeys
alias(object, model, …)
# S4 method for listofkeyrings
alias(object, model, …)
# S4 method for planordesign
alias(object, model, fact, block, …)

Arguments

object

an object of the class.

model

an optional model formula (by default the first model in object) or, when object is a '>keymatrix, a matrix representing factorial model terms.

fact

a character or numeric vector of parent factor names for the columns of object.

block

a logical vector to identify the columns of object associated with a block factor.

…

ignored.

Value

• When object is a '>keymatrix, a vector with (i) the number of unaliased treatment effecs; (ii) the number of mutually aliased treatment effects; (iii) the number of treatment effects aliased with block effects.

• When object is a '>designkey, an invisible NULL.

• When object is a '>listofkeyrings, the factors, the model and the number of solutions for each prime in a list indexed by the primes p of the object. Each element is a 3-column matrix with one row per solution for prime p. The columns give (i) the number of unaliased treatment effecs; (ii) the number of mutually aliased treatment effects; (iii) the number of treatment effects aliased with block effects.

• The method is NOT YET IMPLEMENTED on objects of class '>listofdesignkeys.

• When object is a '>planordesign, this function is the alias method applied on each of the '>keymatrix objects contained in its designkey slot.

Details

When object is a '>keymatrix, “alias” displays the key matrix and the factorial effects confounded with the mean. It prints the unaliased treatment effects, then the groups of aliased treatment effects, then the treatments effects confounded with block effects and finally the unaliased block effects, when considering all the factorial terms that are represented in the model argument, which is set if missing to the identity matrix (main effects only).

Examples

# NOT RUN {
### Creation of an object of class listofkeyrings
K0 <- planor.designkey(factors=c(LETTERS[1:4], "block"), nlevels=rep(3,5),
model=~block+(A+B+C+D)^2, estimate=~A+B+C+D,
nunits=3^3, base=~A+B+C, max.sol=2)
### alias on an object of class keymatrix
alias(K0[[1]][[1]])
### alias on an object of class designkey
alias(K0[1])
### alias on an object of class listofkeyrings
alias(K0)
# }