fwdtr.test.object {Rfwdmv}R Documentation

fwdtr.test.object

Description

An object containing the values of the signed likelihood ratio test estimates of transformation parameters for selected variables and selected values of lambda as specified in function fwdtr.test. The class attribute is set to fwdtr.test.

Format

A fwdtr.test.object is a list with the following elements:

call
the matched call.
test
a list containing a collection of matrices. Each matrix has a number of columns equal to the length of vector lambda.around. The first matrix contains the values of the signed sqrt likelihood ratio test for the first element in vector col.to.compare supplied in function lambda.test.fwdmv. More in detail, the first column contains the values of the signed sqrt likelihood ratio test associated with first element of lambda.around. The second matrix contains the values of the signed sqrt likelihood ratio test for the second element in vector col.to.compare, ecc...
n
an integer, the number of units in the data.
m
an integer, the number of units in the subset in the first step of the forward search.
col.names
a character vector containing the names of the columns of the data set.
col.to.compare
integer vector containing the numbers associated to the variables for which the fan plot has been produced.
lambda.around
Vector specifying for which values of lambda the signed sqrt likelihood ratio has been computed.
message
Message that warn the user on the number of times that optim failed to converge properly. If NULL there are no convergence problem. See optim for details.

Details

Class fwdtr.test object is created by function fwdtr.test.

See Also

fwdtr
computes maximum likelihood estimates of transformation parameters in each step of the forward search.

Examples


data(mussels.dat)

## reasonable values of transformation parameters for the data set
## found using procedure fwdtr

lambda.R <- c(0.5, 0, 0.5, 0, 0)
l.rat <- fwdtr.test(mussels.dat, lambda.R, col.to.compare = 1:5)

## l.rat is a fwdtr.test object

plot.fwdtr.test(l.rat)

[Package Rfwdmv version 0.72-2 Index]