lmdme (version 1.14.0)

leverage: leverage test of lmdme objects

Description

This function calculates the leverage test for each individual using the Principal Component Analysis (comps function) on the coefficients of the given decomposed model term.

Arguments

object

lmdme class object.

comps

a numeric vector indicating the PCA component indexes to keep. Default the first two components (1:2).

term

a character specifying the model term.

level

the quantile level. Default value 0.95

Value

data.frame with the following fields

leverage

numeric for the corresponding row leverage

over

logical indicating if the leverage > quantile(leverage,level) for the given decomposed term

References

Tarazona S, Prado-Lopez S, Dopazo J, Ferrer A, Conesa A, Variable Selection for Multifactorial Genomic Data, Chemometrics and Intelligent Laboratory Systems, 110:113-122 (2012)

See Also

prcomp, quantile

Examples

{
data(stemHypoxia)

##Just to make a balanced dataset in the Fisher sense (2 samples per
## time*oxygen levels)
design<-design[design$time %in% c(0.5, 1, 5) & design$oxygen %in% c(1,5,21),]
design$time<-as.factor(design$time)
design$oxygen<-as.factor(design$oxygen)
rownames(M)<-M[, 1]

##Keeping appropriate samples only
M<-M[, colnames(M) %in% design$samplename]

##ANOVA decomposition
fit<-lmdme(model=~time+oxygen+time:oxygen, data=M, design=design)

##Leverages for the first two Principal Components and q95 (default value).
##Leverages for the first three Principal Components and q99.
leverages2PCDefault<-leverage(fit, term="time:oxygen")
leverages3PCq99<-leverage(fit, comps=1:3, term="time:oxygen", level=0.99)
}