regrresidplot: Draw a plot of y vs x from a linear model object, with...

In HH: Statistical Analysis and Data Display: Heiberger and Holland

Draw a plot of y vs x from a linear model object, with residuals indicated by lines or squares.

Description

Draw a plot of response vector y vs predictor variable x from a linear model object all of whose predictors are a function of x, with residuals indicated by lines or squares.

Usage

regrresidplot(x, y, resid.plot = FALSE, fit.line=TRUE,
              lm.object = lm(y ~ x), x.name = names(lm.object$model)[2],
              col = trellis.par.get()$plot.symbol$col,
              col.yhat = NULL, col.fit = "gray80", col.resid = "gray40", ...)

panel.residSquare(x, y, yhat, resid.plot = FALSE, col = "black", ...)

Arguments

x	Predictor variable. Must be a vector or a single column.
y	Response variable. Must be a vector or a single column.
yhat	Predicted value of y based on the model in lm.object over the xlim range of the plot.
resid.plot	Logical or character. Should the residuals from lm.object be plotted, and how? Default is FALSE. Alternatives are TRUE for lines and "square" for squares.
fit.line	Logical. Should the fitted regression line from lm.object be plotted? Default TRUE.
lm.object	Linear model object of y against some function of x. The default value is the simple linear regression of lm(y ~ x).
x.name	Name of $x$-variable to be used in the construction of the fitted values.
col	Color of observed points.
col.yhat	Color of fitted points. Default is NULL.
col.fit	Color of fitted line.
col.resid	Color of residuals, either lines or squares depending on the value of resid.plot.
...	Additional arguments to the panel functions.

Value

regrresidplot returns a "trellis" object. panel.residSquare is a panel function with no useful returned value.

Author(s)

Richard M. Heiberger <[e-mail]>

Examples

data(fat)
fat.lm <- lm(bodyfat ~ abdomin, data=fat)

AA <- regrresidplot(fat$abdomin, fat$bodyfat, xlim=c(70,185), ylim=c(0,50))
BB <- regrresidplot(fat$abdomin, fat$bodyfat, xlim=c(70,185), ylim=c(0,50),
                    resid.plot="line")
CC <- regrresidplot(fat$abdomin, fat$bodyfat, xlim=c(70,185), ylim=c(0,50),
                    resid.plot="square")

update(between=list(y=1),
       c("Residuals Not Displayed"=AA,
         "Residual Lines"=BB,
         "Residual Squares"=CC, layout=c(1,3)))

HH documentation built on May 29, 2024, 6:24 a.m.