plot.iqr: Plot Quantile Regression Coefficients

Description

Plots quantile regression coefficients \(\beta(p)\) as a function of \(p\), based on a fitted model of class “iqr”.

Usage

# S3 method for iqr
plot(x, conf.int = TRUE, polygon = TRUE, which = NULL, ask = TRUE, ...)

Arguments

x: an object of class “iqr”, typically the result of a call to iqr.
conf.int: logical. If TRUE, asymptotic 95% confidence intervals are added to the plot.
polygon: logical. If TRUE, confidence intervals are represented by shaded areas via polygon. Otherwise, dashed lines are used.
which: an optional numerical vector indicating which coefficient(s) to plot. If which = NULL, all coefficients are plotted.
ask: logical. If which = NULL and ask = TRUE (the default), you will be asked interactively which coefficients to plot.
...: additional graphical parameters, that can include xlim, ylim, xlab, ylab, col, lwd, cex.lab, cex.axis, axes, frame.plot. See par.

Author

Paolo Frumento paolo.frumento@unipi.it

Details

Using iqr, each quantile regression coefficient \(\beta(p)\) is described by a linear combination of known parametric functions of \(p\). With this command, a plot of \(\beta(p)\) versus \(p\) is created. If ask = TRUE, an additional option permits plotting a Q-Q plot of the fitted cumulative distribution function (CDF), that should follow a U(0,1) distribution if the model is correctly specified. If the data are censored or truncated, this is assessed applying the Kaplan-Meier estimator to the fitted CDF values. See also test.fit for a formal test of uniformity.

Examples

Run this code

  
  # using simulated data
  
  n <- 1000
  x <- runif(n)
  qy <- function(p,x){p^2 + x*log(p)}
  # true quantile function: Q(p | x) = beta0(p) + beta1(p)*x, with
     # beta0(p) = p^2
     # beta1(p) = log(p)
  y <- qy(runif(n), x) # to generate y, plug uniform p in qy(p,x)
  
  par(mfrow = c(1,2))
  plot(iqr(y ~ x, formula.p = ~ slp(p,3)), ask = FALSE) 
  # flexible fit with shifted Legendre polynomials