mrlPlot: Diagnostic tools for peaks over threshold analysis

In martindurocher/floodRFA: A R-package for regional frequency of floods

Description

Produce mean residual Life, quantite-quantile, probability-probability, histogram of the excesses, Stability plot for the parameter of the GPA.

Usage

mrlPlot(x, ...)

## S3 method for class 'data.frame'
mrlPlot(obj, ...)

## S3 method for class 'matrix'
mrlPlot(obj, ...)

## S3 method for class 'formula'
mrlPlot(form, x, ...)

## S3 method for class 'numeric'
mrlPlot(x, dt = NULL, u, declust = "run", r = 1,
  rlow = 0.75, alpha = 0.05, ylab = "Mean Residual Life",
  xlab = "Threshold", col = "black", lty = 1, lwd = 1,
  col.ci = "black", lty.ci = 3, lwd.ci = 1, ylim = NULL,
  display = TRUE, ...)

## S3 method for class 'fpot'
hist(obj, main = "", xlab = "Threshold (u)", ...)

## S3 method for class 'fpot'
qqplot(obj, ci = FALSE, nsim = 200, alpha = 0.05,
  xlab = "Theoritical quantiles", ylab = "Empirical quantiles",
  log.scale = TRUE, ...)

## S3 method for class 'fpot'
ppplot(obj, ...)

## S3 method for class 'data.frame'
gpaStabPlot(obj, ...)

## S3 method for class 'matrix'
gpaStabPlot(obj, ...)

## S3 method for class 'formula'
gpaStabPlot(form, x, ...)

## S3 method for class 'numeric'
gpaStabPlot(x, dt = NULL, u, declust = NULL, r = 1,
  rlow = 0.75, alpha = 0.05, param = "scale", method = "mle",
  nboot = 1000, xlab = "Threshold (u)", ylab = NULL, col = "black",
  lty = 1, lwd = 1, col.ci = "black", lty.ci = 3, lwd.ci = 1,
  ylim = NULL, display = TRUE, ...)

Arguments

x, dt, form	Sample and time of observation. A formula can also be passed, in which case x is a data.frame.
...	Others arguments are passed to function plot and which.floodPeaks
u	Series of candidate thresholds.
alpha	Confidence interal with level 1-alpha/2.
param	Choice of the parameter to display. Either 'alpha' or 'kappa'.
method	Estimation method.

Details

The function gpaStabPlot will loop across calls of function fitPot and will produce a graphics of the shape parameter in respect of the threshold u. See the respective functions for a description of the function parameters.

See Also

adTestPlot, dispIndexPlot, fitPot, which.floodPeaks.

Examples

## Find list of candidate thresholds
lstu <- seq(500,2500, len = 100)

mrlPlot(flow~date,canadaFlood$daily, u = lstu, declust = 'flood', r = 14)

gpaStabPlot(flow~date,canadaFlood$daily, u = lstu, declust = 'flood', r = 14)
gpaStabPlot(flow~date,canadaFlood$daily, u = lstu,
            declust = 'flood', r = 14, param = 'alpha', method = 'mle2')


fit <- fitPot(flow~date,canadaFlood$daily, u = 997, declust = 'flood', r = 14)
ppplot(fit)
qqplot(fit, ci = TRUE)
hist(fit)

martindurocher/floodRFA documentation built on June 5, 2019, 8:44 p.m.