agricolae (version 1.2-0)

PBIB.test: Analysis of the Partially Balanced Incomplete Block Design

Description

Analysis of variance PBIB and comparison mean adjusted. Applied to resoluble designs: Lattices and alpha design.

Usage

PBIB.test(block,trt,replication,y,k, method=c("REML","ML","VC"), 
test = c("lsd","tukey"), alpha=0.05, console=FALSE, group=TRUE)

Arguments

block

blocks

trt

Treatment

replication

Replication

y

Response

k

Block size

method

Estimation method: REML, ML and VC

test

Comparison treatments

alpha

Significant test

console

logical, print output

group

logical, groups

Value

• block
  Vector, consecutive numbers by replication
• trt
  Vector numeric or character
• replication
  Vector
• y
  numeric vector
• k
  numeric constant
• method
  Character: REML, ML and VC
• test
  Character: comparison methods lsd and tukey
• alpha
  Numeric
• group
  Logic

Details

Method of comparison treatment. lsd: least significant difference. tukey: Honestly significant differente. Estimate: specifies the estimation method for the covariance parameters. The REML is the default method. The REML specification performs residual (restricted) maximum likelihood, and The ML specification performs maximum likelihood, and the VC specifications apply only to variance component models.

References

1. Iterative Analysis of Generalizad Lattice Designs. E.R. Williams (1977) Austral J. Statistics 19(1) 39-42. 2. Experimental design. Cochran and Cox. Second edition. Wiley Classics Library Edition published 1992

See Also

BIB.test, design.alpha

Examples

require(agricolae)
require(MASS) # method = VC in PBIB.test
# require(nlme) # method = REML or LM in PBIB.test
# alpha design 
Genotype<-paste("geno",1:30,sep="")
ntr<-length(Genotype)
r<-2
k<-3
s<-10
obs<-ntr*r
b <- s*r
book<-design.alpha(Genotype,k,r,seed=5)
book$book[,3]<- gl(20,3)
# dataset
y<-c(5,2,7,6,4,9,7,6,7,9,6,2,1,1,3,2,4,6,7,9,8,7,6,4,3,2,2,1,1,2,
     1,1,2,4,5,6,7,8,6,5,4,3,1,1,2,5,4,2,7,6,6,5,6,4,5,7,6,5,5,4)
dbook<-data.frame(book$book,yield=y)
rm(y,Genotype)
# analysis
require(nlme) # method = REML or LM in PBIB.test
attach(dbook)
model <- PBIB.test(block, Genotype, replication, yield, k=3, method="REML")
detach(dbook)
model$ANOVA
model$method
model$parameters
model$Fstat
head(model$comparison)
model$means
model$groups
bar.group(model$groups,ylim=c(0,9), density=20, las=2)