gppiece: Piece-wise generalized Pareto distribution

In longevity: Statistical Methods for the Analysis of Excess Lifetimes

Piece-wise generalized Pareto distribution

Description

Density, distribution, quantile functions and random number generation from the mixture model of Northrop and Coleman (2014), which consists of m different generalized Pareto distributions over non-overlapping intervals with m shape parameters and one scale parameter; the other scale parameters are constrained so that the resulting distribution is continuous over the domain and reduces to a generalized Pareto distribution if all of the shape parameters are equal.

Usage

dgppiece(x, scale, shape, thresh, log = FALSE)

pgppiece(q, scale, shape, thresh, lower.tail = TRUE, log.p = FALSE)

qgppiece(p, scale, shape, thresh, lower.tail = TRUE, log.p = FALSE)

rgppiece(n, scale, shape, thresh)

Arguments

x, q	vector of quantiles
scale	positive value for the first scale parameter
shape	vector of m shape parameters
thresh	vector of m thresholds
log, log.p	logical; if TRUE, the values are returned on the log scale
lower.tail	logical; if TRUE (default), probabilities are \Pr[X \leq x] otherwise, \Pr[X > x].
p	vector of probabilities
n	sample size

Value

a vector of quantiles (qgppiece), probabilities (pgppiece), density (dgppiece) or random number generated from the model (rgppiece)

References

Northrop & Coleman (2014). Improved threshold diagnostic plots for extreme value analyses, Extremes, 17(2), 289–303.

longevity documentation built on Nov. 12, 2023, 5:07 p.m.