| % File src/library/stats/man/lm.summaries.Rd |
| % Part of the R package, https://www.R-project.org |
| % Copyright 1995-2011 R Core Team |
| % Distributed under GPL 2 or later |
| |
| \name{lm.summaries} |
| \alias{family.lm} |
| \alias{formula.lm} |
| \alias{residuals.lm} |
| \alias{labels.lm} |
| \title{Accessing Linear Model Fits} |
| \usage{ |
| \method{family}{lm}(object, \dots) |
| |
| \method{formula}{lm}(x, \dots) |
| |
| \method{residuals}{lm}(object, |
| type = c("working", "response", "deviance", "pearson", |
| "partial"), |
| \dots) |
| |
| \method{labels}{lm}(object, \dots) |
| } |
| \arguments{ |
| \item{object, x}{an object inheriting from class \code{lm}, usually |
| the result of a call to \code{\link{lm}} or \code{\link{aov}}.} |
| \item{\dots}{further arguments passed to or from other methods.} |
| \item{type}{the type of residuals which should be returned. Can be abbreviated.} |
| } |
| \description{ |
| All these functions are \code{\link{methods}} for class \code{"lm"} objects. |
| } |
| \details{ |
| The generic accessor functions \code{coef}, \code{effects}, |
| \code{fitted} and \code{residuals} can be used to extract |
| various useful features of the value returned by \code{lm}. |
| |
| The working and response residuals are \sQuote{observed - fitted}. The |
| deviance and pearson residuals are weighted residuals, scaled by the |
| square root of the weights used in fitting. The partial residuals |
| are a matrix with each column formed by omitting a term from the |
| model. In all these, zero weight cases are never omitted (as opposed |
| to the standardized \code{\link{rstudent}} residuals, and the |
| \code{\link{weighted.residuals}}). |
| |
| How \code{residuals} treats cases with missing values in the original |
| fit is determined by the \code{na.action} argument of that fit. |
| If \code{na.action = na.omit} omitted cases will not appear in the |
| residuals, whereas if \code{na.action = na.exclude} they will appear, |
| with residual value \code{NA}. See also \code{\link{naresid}}. |
| |
| The \code{"lm"} method for generic \code{\link{labels}} returns the |
| term labels for estimable terms, that is the names of the terms with |
| an least one estimable coefficient. |
| } |
| \seealso{ |
| The model fitting function \code{\link{lm}}, \code{\link{anova.lm}}. |
| |
| \code{\link{coef}}, \code{\link{deviance}}, |
| \code{\link{df.residual}}, |
| \code{\link{effects}}, \code{\link{fitted}}, |
| \code{\link{glm}} for \bold{generalized} linear models, |
| \code{\link{influence}} (etc on that page) for regression diagnostics, |
| \code{\link{weighted.residuals}}, |
| \code{\link{residuals}}, \code{\link{residuals.glm}}, |
| \code{\link{summary.lm}}, \code{\link{weights}}. |
| |
| \link{influence.measures} for deletion diagnostics, including |
| standardized (\code{\link{rstandard}}) |
| and studentized (\code{\link{rstudent}}) residuals. |
| } |
| \references{ |
| Chambers, J. M. (1992) |
| \emph{Linear models.} |
| Chapter 4 of \emph{Statistical Models in S} |
| eds J. M. Chambers and T. J. Hastie, Wadsworth & Brooks/Cole. |
| } |
| \examples{ |
| \dontshow{utils::example("lm", echo = FALSE)} |
| ##-- Continuing the lm(.) example: |
| coef(lm.D90) # the bare coefficients |
| |
| ## The 2 basic regression diagnostic plots [plot.lm(.) is preferred] |
| plot(resid(lm.D90), fitted(lm.D90)) # Tukey-Anscombe's |
| abline(h = 0, lty = 2, col = "gray") |
| |
| qqnorm(residuals(lm.D90)) |
| } |
| \keyword{regression} |
| \keyword{models} |