glmx: pregibon – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

pregibon

Pregibon Distribution

Description

Density, distribution function, quantile function and random generation for the Pregibon distribution with parameters a and b. It is a special case of the generalized Tukey lambda distribution.

Usage

dpregibon(x, a = 0, b = 0, log = FALSE, tol = 1e-12)
ppregibon(q, a = 0, b = 0, lower.tail = TRUE, log.p = FALSE, tol = 1e-12)
qpregibon(p, a = 0, b = 0, lower.tail = TRUE, log.p = FALSE)
rpregibon(n, a = 0, b = 0)

Arguments

`x, q`	vector of quantiles.
`p`	vector of probabilities.
`n`	number of observations. If `length(n) > 1`, the length is taken to be the number required.
`a, b`	distribution parameters.
`log, log.p`	logical; if TRUE, probabilities p are given as log(p).
`lower.tail`	logical; if TRUE (default), probabilities are P[X ≤ x], otherwise, P[X > x].
`tol`	numeric tolerance for computation of the distribution function.

Details

The distribution is a special case of the generalized Tukey lambda distribution and is used by Pregibon (1980) for goodness-of-link testing. See Koenker (2006) and Koenker and Yoon (2009) for more details.

The implementation is based on the corresponding functions for the GeneralisedLambdaDistribution in the gld package (King 2013).

The corresponding link generator is available in the function pregibon.

Value

dpregibon gives the probability density function, ppregibon gives the cumulative distribution function, qpregibon gives the quantile function, and rpregibon generates random deviates.

References

King R (2013). “Estimation and Use of the Generalised (Tukey) Lambda Distribution.” R package version 2.0.1. http://CRAN.R-project.org/package=gld

Koenker R (2006). “Parametric Links for Binary Response.” R News, 6(4), 32–34.

Koenker R, Yoon J (2009). “Parametric Links for Binary Choice Models: A Fisherian-Bayesian Colloquy.” Journal of Econometrics, 152, 120–130.

Pregibon D (1980). “Goodness of Link Tests for Generalized Linear Models.” Journal of the Royal Statistical Society C, 29, 15–23.

Examples

## Koenker & Yoon (2009), Figure 2
par(mfrow = c(3, 3))
pregiboncurve <- function(a, b, from, to, n = 301) {
  dp <- function(x) dpregibon(x, a = a, b = b)
  curve(dp, from = from, to = to, n = n,
    xlab = "", ylab = "",
    main = paste("a = ", a, ", b = ", b, sep = ""))
}
pregiboncurve(-0.25, -0.25,  -5, 65)
pregiboncurve(-0.25,  0,    -18, 18)
pregiboncurve(-0.25,  0.25, -65,  5)
pregiboncurve( 0,    -0.25,  -4, 22)
pregiboncurve( 0,     0,     -8,  8)
pregiboncurve( 0,     0.25, -22,  4)
pregiboncurve( 0.25, -0.25,  -2.4,9)
pregiboncurve( 0.25,  0,     -4,  4)
pregiboncurve( 0.25,  0.25,  -9,  2.4)
par(mfrow = c(1, 1))

glmx

Generalized Linear Models Extended

v0.1-1

GPL-2 | GPL-3

Authors

Achim Zeileis [aut, cre], Roger Koenker [aut], Philipp Doebler [aut]

Initial release

2015-11-19