adTest: Anderson-Daring goodness-of-fit test

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

Description

Perform the Anderson-darling goodness-of-fit test as well of the modified versions of the test, which puts more weight to the lower or upper tails of the distribution. The null hypothesis is that the data are generated by a given distribution (composite). P-values are computed by parametric bootstrap.

Usage

adTest(obj, ...)

## Default S3 method:
adTest(x, type, method = "lmom", nboot = 1000,
  para = NULL, bsample = FALSE, ...)

## S3 method for class 'atSite'
adTest(obj, nboot = 1000, bsample = FALSE, cores = NULL)

## S3 method for class 'fpot'
adTest(obj, method = "tab", nboot = 1000, bsample = FALSE,
  cores = NULL)

Arguments

x, obj	Vector of data or output of function atSite.
type	Distribution of the null hypothesis. See vec2par.
method	Estimation method.
nboot	Number of bootstrap sample for calculating the p-value.
para	An object containing an estimation of the distribution. See vec2par (optional).
bsample	Logical. Should the bootstrap sample be returned.
cores	Number of cores to use for parallel computing.

References

Heo, J.-H., Shin, H., Nam, W., Om, J., & Jeong, C. (2013). Approximation of modified Anderson–Darling test statistics for extreme value distributions with unknown shape parameter. Journal of Hydrology, 499, 41–49. https://doi.org/10.1016/j.jhydrol.2013.06.008

Examples

library(lmomco)

x0 <- rlmomco(50,vec2par(c(100,5),'nor'))

adTest(x0, 'gum', nboot = 100)

print(ax <- atSite(x0))
adTest(ax)

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