swh:1:snp:a3b81a63bc885946c7e57d5f25f804425977c98d
Tip revision: 5ad24cff87e41a020a678cf6e742678aa2ee9dce authored by Roger Koenker on 29 May 2018, 10:33:02 UTC
version 5.36
version 5.36
Tip revision: 5ad24cf
rq.object.Rd
\name{rq.object}
\alias{rq.object}
\alias{formula.rq}
\alias{logLik.rq}
\alias{logLik.rqs}
\alias{AIC.rq}
\alias{AIC.rqs}
\alias{extractAIC.rq}
\title{
Linear Quantile Regression Object
}
\description{
These are objects of class \code{"rq"}.
They represent the fit of a linear conditional quantile function model.
}
\section{Generation}{
This class of objects is returned from the \code{rq} function
to represent a fitted linear quantile regression model.
}
\section{Methods}{
The \code{"rq"} class of objects has methods for the following generic
functions:
\code{coef}, \code{effects}
, \code{formula}
, \code{labels}
, \code{model.frame}
, \code{model.matrix}
, \code{plot}
, \code{logLik}
, \code{AIC}
, \code{extractAIC}
, \code{predict}
, \code{print}
, \code{print.summary}
, \code{residuals}
, \code{summary}
}
\section{Structure}{
The following components must be included in a legitimate \code{rq} object.
\describe{
\item{\code{coefficients}}{
the coefficients of the quantile regression fit.
The names of the coefficients are the names of the
single-degree-of-freedom effects (the columns of the
model matrix).
If the model was fitted by method \code{"br"} with \code{ci=TRUE}, then
the coefficient component consists of a matrix whose
first column consists of the vector of estimated coefficients
and the second and third columns are the lower and upper
limits of a confidence interval for the respective coefficients.
}
\item{\code{residuals}}{
the residuals from the fit.
}
\item{\code{dual}}{
the vector dual variables from the fit.
}
\item{\code{rho}}{
The value(s) of objective function at the solution.
}
\item{\code{contrasts}}{
a list containing sufficient information to construct the contrasts
used to fit any factors occurring in the model.
The list contains entries that are either matrices or character vectors.
When a factor is coded by contrasts, the corresponding contrast matrix
is stored in this list.
Factors that appear only as dummy variables and variables in the model
that are matrices correspond to character vectors in the list.
The character vector has the level names for a factor or the column
labels for a matrix.
}
\item{\code{model}}{
optionally the model frame, if \code{model=TRUE}.
}
\item{\code{x}}{
optionally the model matrix, if \code{x=TRUE}.
}
\item{\code{y}}{
optionally the response, if \code{y=TRUE}.
}
}
}
\details{
The coefficients, residuals, and effects may be extracted
by the generic functions of the same name, rather than
by the \code{$} operator. For pure \code{rq} objects this is less critical
than for some of the inheritor classes. In particular, for fitted rq objects
using "lasso" and "scad" penalties, \code{logLik} and \code{AIC} functions
compute degrees of freedom of the fitted model as the number of estimated
parameters whose absolute value exceeds a threshold \code{edfThresh}. By
default this threshold is 0.0001, but this can be passed via the \code{AIC}
function if this value is deemed unsatisfactory. The function \code{AIC}
is a generic function in R, with parameter \code{k} that controls the form
of the penalty: the default value of \code{k} is 2 which yields the classical
Akaike form of the penalty, while \code{k <= 0} yields the Schwarz (BIC)
form of the penalty.
Note that the extractor function \code{coef} returns a vector with missing values
omitted.
}
\seealso{
\code{\link{rq}}, \code{\link{coefficients}}.
}
\keyword{regression}