% File src/library/stats/man/termplot.Rd % Part of the R package, https://www.R-project.org % Copyright 1995-2021 R Core Team % Distributed under GPL 2 or later \name{termplot} \alias{termplot} \title{Plot Regression Terms} \description{ Plots regression terms against their predictors, optionally with standard errors and partial residuals added. } \usage{ termplot(model, data = NULL, envir = environment(formula(model)), partial.resid = FALSE, rug = FALSE, terms = NULL, se = FALSE, xlabs = NULL, ylabs = NULL, main = NULL, col.term = 2, lwd.term = 1.5, col.se = "orange", lty.se = 2, lwd.se = 1, col.res = "gray", cex = 1, pch = par("pch"), col.smth = "darkred", lty.smth = 2, span.smth = 2/3, ask = dev.interactive() && nb.fig < n.tms, use.factor.levels = TRUE, smooth = NULL, ylim = "common", plot = TRUE, transform.x = FALSE, \dots) } \arguments{ \item{model}{fitted model object} \item{data}{data frame in which variables in \code{model} can be found} \item{envir}{environment in which variables in \code{model} can be found} \item{partial.resid}{logical; should partial residuals be plotted?} \item{rug}{add \link{rug}plots (jittered 1-d histograms) to the axes?} \item{terms}{which terms to plot (default \code{NULL} means all terms); a vector passed to \code{\link{predict}(.., type = "terms", terms = *)}.} \item{se}{plot pointwise standard errors?} \item{xlabs}{vector of labels for the x axes} \item{ylabs}{vector of labels for the y axes} \item{main}{logical, or vector of main titles; if \code{TRUE}, the model's call is taken as main title, \code{NULL} or \code{FALSE} mean no titles.} \item{col.term, lwd.term}{color and line width for the \sQuote{term curve}, see \code{\link{lines}}.} \item{col.se, lty.se, lwd.se}{color, line type and line width for the \sQuote{twice-standard-error curve} when \code{se = TRUE}.} \item{col.res, cex, pch}{color, plotting character expansion and type for partial residuals, when \code{partial.resid = TRUE}, see \code{\link{points}}.} \item{ask}{logical; if \code{TRUE}, the user is \emph{ask}ed before each plot, see \code{\link{par}(ask=.)}.} \item{use.factor.levels}{Should x-axis ticks use factor levels or numbers for factor terms?} \item{smooth}{\code{NULL} or a function with the same arguments as \code{\link{panel.smooth}} to draw a smooth through the partial residuals for non-factor terms} \item{lty.smth, col.smth, span.smth}{Passed to \code{smooth}} \item{ylim}{an optional range for the y axis, or \code{"common"} when a range sufficient for all the plot will be computed, or \code{"free"} when limits are computed for each plot.} \item{plot}{if set to \code{FALSE} plots are not produced: instead a list is returned containing the data that would have been plotted.} \item{transform.x}{logical vector; if an element (recycled as necessary) is \code{TRUE}, partial residuals for the corresponding term are plotted against transformed values. The model response is then a straight line, allowing a ready comparison against the data or against the curve obtained from \code{smooth-panel.smooth}.} \item{\dots}{other graphical parameters.} } \details{ The \code{model} object must have a \code{predict} method that accepts \code{type = "terms"}, e.g., \code{\link{glm}} in the \pkg{stats} package, \code{\link[survival]{coxph}} and \code{\link[survival]{survreg}} in the \CRANpkg{survival} package. For the \code{partial.resid = TRUE} option \code{model} must have a \code{\link{residuals}} method that accepts \code{type = "partial"}, which \code{\link{lm}} and \code{\link{glm}} do. The \code{data} argument should rarely be needed, but in some cases \code{termplot} may be unable to reconstruct the original data frame. Using \code{na.action=na.exclude} makes these problems less likely. Nothing sensible happens for interaction terms, and they may cause errors. The \code{plot = FALSE} option is useful when some special action is needed, e.g.\sspace{}to overlay the results of two different models or to plot confidence bands. } \value{ For \code{plot = FALSE}, a list with one element for each plot which would have been produced. Each element of the list is a data frame with variables \code{x}, \code{y}, and optionally the pointwise standard errors \code{se}. For continuous predictors \code{x} will contain the ordered unique values and for a factor it will be a factor containing one instance of each level. The list has attribute \code{"constant"} copied from the predicted terms object. Otherwise, the number of terms, invisibly. } \seealso{For (generalized) linear models, \code{\link{plot.lm}} and \code{\link{predict.glm}}.} \examples{ require(graphics) had.splines <- "package:splines" \%in\% search() if(!had.splines) rs <- require(splines) x <- 1:100 z <- factor(rep(LETTERS[1:4], 25)) y <- rnorm(100, sin(x/10)+as.numeric(z)) model <- glm(y ~ ns(x, 6) + z) par(mfrow = c(2,2)) ## 2 x 2 plots for same model : termplot(model, main = paste("termplot( ", deparse(model$call)," ...)")) termplot(model, rug = TRUE) termplot(model, partial.resid = TRUE, se = TRUE, main = TRUE) termplot(model, partial.resid = TRUE, smooth = panel.smooth, span.smth = 1/4) if(!had.splines && rs) detach("package:splines") if(requireNamespace("MASS", quietly = TRUE)) { hills.lm <- lm(log(time) ~ log(climb)+log(dist), data = MASS::hills) termplot(hills.lm, partial.resid = TRUE, smooth = panel.smooth, terms = "log(dist)", main = "Original") termplot(hills.lm, transform.x = TRUE, partial.resid = TRUE, smooth = panel.smooth, terms = "log(dist)", main = "Transformed") }} \keyword{hplot} \keyword{regression}