makeContrMat: Make Contrast Matrix

In infoDecompuTE: Information Decomposition of Two-Phase Experiments

Description

Construct a list of contrast matrices for block for treatment effects.

Usage

makeContrMat(design.df, effectNames, effectsMatrix, contr.vec)

Arguments

design.df	a data frame containing the experimental design. Requires every column be a factor.
effectNames	a vector of character containing the labels of the treatment or block terms in the model generated by the terms.
effectsMatrix	a matrix of variables by terms showing which variables appear in which terms generated by the terms.
contr.vec	a list of contrast vectors, this allows the user to specify the contrasts for each treatment or block factor. Note that if this argument is used, it is necessary to specify the contrasts for every treatment or block factor with the same order as effectNames. Default is NA, and the function output the C matrices described by John and Williams (1987).

Details

The main purpose of this function is to compute a list of C matrices described by John and Williams (1987). These C matrices are used for the information decomposition for every treatment effect in every stratum of the experiment.

If the user input their own defined contrasts for each treatment effects. This function will then transform the input contrasts to the C matrices for the treatment effects.

For the two-phase experiments, the same method of information decomposition is used for the block effects of Phase 1 experiment in the stratum defined from the block structure of the Phase 2 experiment. Hence, the C matrices for the block effects of Phase 1 experiment can also be constructed using this function.

Value

A list of contrast matrices.

Author(s)

Kevin Chang

References

John J, Williams E (1987). Cyclic and computer generated Designs. Second edition. Chapman & Hall.

Examples

design1 <- local({ 
  Ani = as.factor(LETTERS[c(1,2,3,4,
                            5,6,7,8)])
  Trt = as.factor(letters[c(1,1,1,1,
                            2,2,2,2)])
  data.frame(Ani, Trt, stringsAsFactors = TRUE )
})

trt.str = "Trt"              
fT <- terms(as.formula(paste("~", trt.str, sep = "")), keep.order = TRUE)  #fixed terms

trtTerm <- attr(fT, "term.labels")
effectsMatrix <- attr(fT, "factor")        

T <- makeContrMat(design1, trtTerm, effectsMatrix, contr.vec = NA)
		
#Fit each treatment contrasts as a vector seperately
Trt1 <- rep(c(1,-1), each = 4)
Trt2 <-  rep(c(1,-1), time = 4)
Trt3 <- Trt1*Trt2
  
T <- makeContrMat(design1, trtTerm, effectsMatrix, 
      contr.vec =list(Trt = list(Trt1 = Trt1, Trt2 = Trt2, Trt3 = Trt3)))

infoDecompuTE documentation built on April 14, 2020, 7:08 p.m.