add_regres_line: Add a regression line and confidence band to a plot

In RemkoDuursma/nlshelper: Convenient Functions for Non-Linear Regression

Description

Plots a regression line from a simple linear model (of the form lm(y ~ x)) to a plot. Also plots the confidence band for the mean, which is calculated using predict.lm.

Usage

add_regres_line(fit, from = NULL, to = NULL, band = TRUE,
  ci.col = "#BEBEBEB3", ...)

Arguments

fit	Object returned by lm. Only models of the form y ~ x are supported, without expressions in I() (see Examples), or interactions, or multiple variables.
from	Optional (read from fitted model); Draw from this X value.
to	Optional (read from fitted model); Draw to this x value.
band	Logical. Whether to add a confidence band.
ci.col	Colour of the confidence band, if plotted. Defaults to a transparent grey colour.
...	Further arguments passed to abline_range

Examples

#'Add a line across the range of the data from a regression object
with(mtcars, plot(1/wt, mpg, xlim=c(0,0.8), ylim=c(0,40)))

# add_regres_line does not allow I() expressions; yet.
mtcars$inv_wt <- 1 / mtcars$wt
fit <- lm(mpg ~ inv_wt, data=mtcars)
add_regres_line(fit)

# Add the regression line and confidence band behind the data
fit <- lm(height ~ age, data=Loblolly)
with(Loblolly, plot(age, height, pch=19, panel.first=add_regres_line(fit)))

RemkoDuursma/nlshelper documentation built on May 9, 2019, 9:39 a.m.