fwdtrProfilePlot {Rfwdmv}R Documentation

Plot method for profile.fwdtr objects

Description

Produces a plot of profile loglikelihoods of transformation parameters in a particular step of the forward search

Usage

fwdtrProfilePlot(x, psfrag.labels = FALSE)

Arguments

x a profile.fwdtr object.
psfrag.labels a logical value. If TRUE then the x, y, and main labels are set to "xlab", "ylab", and "main" for replacement via the psfrag utility.

Details

This plot shows which variables show a sharp definition of the estimates of the transformation parameters and which are the variables whose value of lambda is not very well determined.

Value

an empty list is invisibly returned.

Author(s)

Fabrizio Laurini

References

Atkinson, A. C., Riani, M. and Cerioli, A. (2004) Exploring Multivariate Data with the Forward Search. Springer-Verlag New York.

See Also

fwdtr.object

Examples

data(mussels.dat)

## Forward search on untransformed data 
## Compute max lik. estimates of tramsformation parameters

l.mle <- fwdtr(mussels.dat)

## Compute profile loglikelihoods for transformation parameters in the
## last step of the search and create an object of class profile.fwdtr

l.profile.mle <- profile.fwdtr(l.mle)

## plot the profile loglikelihoods of transformation parameters for
## each variable 

fwdtrProfilePlot(l.profile.mle)

## Test variables 2 and 5
## The forward is based on untransformed data for variables 1, 3 and 4
## sqrt for variable 2 and third root for variable 5

l.mle <- fwdtr(mussels.dat, lambda = c(0.5, 1/3), col.to.transform = c(2, 5))

## build profile likelihood for transformed variables.
## Profile function takes into account that we have also untransformed variables.

l.profile.mle <- profile.fwdtr(l.mle)
fwdtrProfilePlot(l.profile.mle)

[Package Rfwdmv version 0.72-2 Index]