Cooley19: Derives bivariate isolines using the non-parametric approach...

In rjaneUCF/MultiHazard: Tools for modeling compound events

Derives bivariate isolines using the non-parametric approach of Cooley et al. (2019).

Description

The Cooley et al. (2019) method exploits bivariate regular variation and kernel density estimation to generate isolines of bivariate exceedance probabilities. The function utilizes the ks and texmex packages, and works for both asymptotic dependence and independence.

Usage

Cooley19(Data, Migpd, p.base = 0.01, p.proj = 0.001, u = 0.95,
  PLOT = FALSE, x_lim_min_T = NA, x_lim_max_T = NA,
  y_lim_min_T = NA, y_lim_max_T = NA, x_lim_min = NA,
  x_lim_max = NA, y_lim_min = NA, y_lim_max = NA)

Arguments

Data	Data frame consisting of two columns.
Migpd	An Migpd object, containing the generalized Pareto models fitted (independently) to the variables comprising the columns of Data.
p.base	Numeric vector of length one specifying the exceedance probability of the base isoline. Default is 0.01.
p.proj	Numeric vector of length one specifying the exceedance probability of the projected isoline. Default is 0.001.
u	Numeric vector of length one specifying the quantile at which to estimate the asymptotic nature of the data i.e. chi and chibar. Default is 0.95.
PLOT	Logical; indicating whether to plot the base and projected isolines on the original and transformed scale. Default is FALSE.
x_lim_min_T	Numeric vector of length one specifying the lower x-axis limit of the transformed scale plot. Default is NA.
x_lim_max_T	Numeric vector of length one specifying the upper x-axis limit of the transformed scale plot. Default is NA.
y_lim_min_T	Numeric vector of length one specifying the lower y-axis limit of the transformed scale plot. Default is NA.
y_lim_max_T	Numeric vector of length one specifying the upper y-axis limit of the transformed scale plot. Default is NA.
x_lim_min	Numeric vector of length one specifying the lower x-axis limit of the plot on the original scale. Default is NA.
x_lim_max	Numeric vector of length one specifying the upper x-axis limit of the plot on the original scale. Default is NA.
y_lim_min	Numeric vector of length one specifying the lower y-axis limit of the plot on the original scale. Default is NA.
y_lim_max	Numeric vector of length one specifying the lower y-axis limit of the plot on the original scale. Default is NA.

Value

List comprising a description of the type of (asymptoptic) dependence Asym, the values the extremal dependence measures Chi and n.bar, exceedance probabilities of the base p.base and projected p.proj isolines, as well as the points on the base I.base and projected I.proj isolines.

See Also

Dataframe_Combine Decluster GPD_Fit Migpd_Fit

Examples

S20.GPD<-Migpd_Fit(Data=S20.Detrend.Declustered.df[,-1], mqu =c(0.99,0.99,0.99))
Cooley19(Data=na.omit(S20.Detrend.df[,3:4]),Migpd=s.Migpd,
p.base=0.01,p.proj=0.001,PLOT=TRUE,x_lim_max_T=500,y_lim_max_T=500)

rjaneUCF/MultiHazard documentation built on April 20, 2024, 12:48 a.m.