R/linefit.R

In ejanalysis/analyze.stuff: Miscellaneous Tools for Analyzing Data in Rows and Columns

#' @title Add fit lines to a scatter plot
#'
#' @description
#' Convenient wrapper for stats::lowess(), stats::lm(), and stats::coef(stats::line())
#'
#' @details
#' This function adds lines to a scatter plot, using
#' lines([stats::lowess()](x,y)),
#' abline([stats::lm()](y~x)), and
#' abline([stats::coef()]([stats::line()](x,y)))
#' DOESN'T SEEM TO WORK IF log = 'xy' was used in original plot()
#' NOTE: coef(line())  and lm()  give different results
#' @param x x values, required
#' @param y y values, required
#' @param type passed through to [graphics::lines()] for the [stats::lowess()]
#' @param cex scaling for lowess
#' @param show.lowess Logical value, optional, TRUE by default. Defines if lowess is shown
#' @param show.lm Logical value, optional, TRUE by default. Defines if lm line is shown
#' @param show.line Logical value, optional, TRUE by default. Defines if should show abline(coef(line(x,y)))
#' @return Provides a plot just as a side effect
#' @examples
#' \dontrun{
#'     # see
#' #?lm  or  ?aov   or  ?glm
#' # ?line
#' require(graphics)
#' plot(cars)
#' (z <- line(cars))
#' abline(coef(z))
#' ## Tukey-Anscombe Plot :
#' plot(residuals(z) ~ fitted(z), main = deparse(z$call))
#' # ?predict
#' # ?lowess
#' # ?scatterplot
#' #The scatterplot( ) function in the car package offers many enhanced features, including
#' #fit lines, marginal box plots, conditioning on a factor, and interactive point identification.
#' #Each of these features is optional.
#' # Enhanced Scatterplot of MPG vs. Weight
#' # by Number of Car Cylinders
#'  library(car)
#'  scatterplot(mpg ~ wt | cyl, data = mtcars,
#'              xlab = "Weight of Car", ylab = "Miles Per Gallon",
#'              main = "Enhanced Scatter Plot",
#'              labels = row.names(mtcars))
#' }
#' @export
linefit <- function(x, y, type = 'b', cex = 4, show.lowess = TRUE, show.lm = TRUE, show.line = TRUE) {
  if (show.lowess) { lines(lowess(x, y), type = type, col = "blue", pch = '.', cex = cex) } # lowess line (x, y)
  if (show.lm)     { abline(lm(y ~ x, na.action = na.exclude), col = "dark green") } # regression line (y ~ x)
  if (show.line)   { abline(coef(line(x,y)), col = 'light green') }
}