The R Project SVN R

% File src/library/stats/man/confint.Rd
% Part of the R package, https://www.R-project.org
% Copyright 1995-2025 R Core Team
% Distributed under GPL 2 or later

\name{confint}
\title{Confidence Intervals for Model Parameters}
\alias{confint}
\alias{confint.default}
\alias{confint.lm}
\alias{confint.glm}
\alias{confint.nls}
\description{
  Computes confidence intervals for one or more parameters in a fitted
  model.  There is a default and a method for objects inheriting from class
  \code{"\link{lm}"} and several other classes.
}
\usage{
confint(object, parm, level = 0.95, \dots)

\S3method{confint}{default}(object, parm, level = 0.95, \dots)
\S3method{confint}{lm}(object, parm, level = 0.95, \dots)
\S3method{confint}{glm}(object, parm, level = 0.95, trace = FALSE, test=c("LRT", "Rao"), \dots)
\S3method{confint}{nls}(object, parm, level = 0.95, \dots)
}
\arguments{
  \item{object}{a fitted model object.}
  \item{parm}{a specification of which parameters are to be given
    confidence intervals, either a vector of numbers or a vector of
    names.  If missing, all parameters are considered.}
  \item{level}{the confidence level required.}
  \item{trace}{ logical.  Should profiling be traced?}
  \item{test}{use Likelihood Ratio or \I{Rao} Score test in profiling.}
  \item{\dots}{additional argument(s) for methods.}
}
\value{
  A matrix (or vector) with columns giving lower and upper confidence
  limits for each parameter. These will be labelled as (1-level)/2 and
  1 - (1-level)/2 in \% (by default 2.5\% and 97.5\%).
}
\details{
  \code{confint} is a generic function.  The default method is based on
  inverting the Wald test, assuming approximate
  normality of the estimator.  It needs suitable \code{\link{coef}} and
  \code{\link{vcov}} methods to be available.  The default method can be
  called directly for comparison with other methods.  It also works for 
  fitted models in the S4 class system.

  For objects of class \code{"lm"} the direct formulae based on \eqn{t}
  values are used.

  Methods for classes \code{"glm"}
  and \code{"nls"} call the appropriate profile method,
  then find the confidence intervals by interpolation in the profile
  traces.  If the profile object is already available it can be used
  as the main argument rather than the fitted model object itself.
}
\references{
  \bibshow{R:Venables+Ripley:2002}
}
\seealso{
  Original versions: \code{\link[MASS]{confint.glm}} and
  \code{\link[MASS]{confint.nls}} in package \CRANpkg{MASS}.
}
\examples{
fit <- lm(100/mpg ~ disp + hp + wt + am, data = mtcars)
confint(fit)
confint(fit, "wt")

## from example(glm)
counts <- c(18,17,15,20,10,20,25,13,12)
outcome <- gl(3, 1, 9); treatment <- gl(3, 3)
glm.D93 <- glm(counts ~ outcome + treatment, family = poisson())
confint(glm.D93) 
confint.default(glm.D93)  # based on asymptotic normality}
\keyword{models}