fwdtrProfilePlot {Rfwdmv} | R Documentation |
Produces a plot of profile loglikelihoods of transformation parameters in a particular step of the forward search
fwdtrProfilePlot(x, psfrag.labels = FALSE)
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. |
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.
an empty list is invisibly returned.
Fabrizio Laurini
Atkinson, A. C., Riani, M. and Cerioli, A. (2004) Exploring Multivariate Data with the Forward Search. Springer-Verlag New York.
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)