infoDecompMat: Construct the Matrix from Information Decomposition

Description Usage Arguments Details Value Author(s) Examples

View source: R/infoDecompMat.R

Description

Perform the information decomposition for either the block or treatment effects within a single stratum.

Usage

1
infoDecompMat(z, T, N)

Arguments

z

a matrix containing the orthogonal projector for a single stratum generated by makeOrthProjectors.

T

a list of contrast matrices generated by makeContrMat.

N

a matrix containing the design matrix generated by makeOverDesMat.

Details

The main purpose of this function is to construct a list of resultant matrices associated with each source of variation after the information decomposition.

This list of matrices are then used to compute the coefficient of the variance components in the expected mean squares.

Value

A list of matrices.

Author(s)

Kevin Chang

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
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)
})

blk.str = "Ani"
    
rT = terms(as.formula(paste("~", blk.str, sep = "")), keep.order = TRUE) 
blkTerm = attr(rT,"term.labels")
     
Z = makeBlkDesMat(design1, blkTerm)
Pb = makeOrthProjectors(Z)

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

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

T <- makeContrMat(design1, trtTerm, effectsMatrix, contr.vec = NA)

N =  makeOverDesMat(design1, trtTerm)


infoDecompMat(Pb[[1]], T, N)

infoDecompuTE documentation built on May 29, 2018, 9:05 a.m.