goft (version 1.3.6)

ev_test: Tests for the extreme value distributions

Description

Correlation and ratio tests for Fisher-Tippet extreme value distributions.

Usage

ev_test(x, dist = "gumbel", method = "cor", N = 1000)

Arguments

x

a numeric data vector containing a random sample.

dist

the extreme value distribution to be tested. Options are "gumbel", "frechet" and "weibull" for testing the Gumbel, Frechet and Weibull distribution hypotheses. Default option is "gumbel".

method

the test to be used. Available options are "ratio" and "cor". Default is "cor".

N

number of Monte Carlo samples used to approximate the p-value of the test when "ratio" option is chosen. Default is N = 1000.

Value

A list with class "htest" containing the following components.

statistic

the calculated value of the test statistic.

p.value

the approximated p-value of the test.

method

a character string for the method used to test the null hypothesis.

data.name

a character string giving the name of the data set.

Details

Option "ratio" performs a test based on the ratio of two estimators for the variance of the Gumbel (type I extreme value) distribution (Gonzalez-Estrada and Villasenor, 2018).

Option "cor" performs a test based on the max-stability property of extreme value distributions (Gonzalez-Estrada and Villasenor, 2010). The sample size must lie between 20 and 250.

References

Gonzalez-Estrada, E. and Villasenor, J.A. (2018). An R package for testing goodness of fit: goft. Journal of Statistical Computation and Simulation, 88 4, 726-751. https://doi.org/10.1080/00949655.2017.1404604

Gonzalez-Estrada, E. and Villasenor-Alva, J.A. (2010). A Goodness-of-Fit Test for Location-Scale Max-Stable Distributions. Communications in Statistics: Simulation and Computation, 39 3, 557-562. https://doi.org/10.1080/03610910903528293

Examples

# NOT RUN {
# Example 1:  testing the Gumbel distribution hypothesis using the correlation test
x <- -log(rexp(20))   # simulating a data set from the standard Gumbel distribution
ev_test(x)            

# }