gof.bootstrap: Generic GOF tests based on EDF using bootstrap

In EDFtest: Goodness of Fit Based on Empirical Distribution Function

Description

This function takes in an i.i.d. random sample, use MLE to estimate parameters of the assumed distribution, compute probability integral transforms, and computes Cramér-von Mises, Anderson-Darling and Watson statistics and their P-values using bootstrap method.

Usage

gof.uniform.bootstrap(x, M = 10000)

gof.normal.bootstrap(x, M = 10000)

gof.gamma.bootstrap(x, M = 10000)

gof.logistic.bootstrap(x, M = 10000)

gof.laplace.bootstrap(x, M = 10000)

gof.weibull.bootstrap(x, M = 10000)

gof.extremevalue.bootstrap(x, M = 10000)

gof.exp.bootstrap(x, M = 10000)

Arguments

x	A random sample.
M	Number of bootstrap, 10000 by default.

Value

Cramér-von Mises, Anderson-Darling and Watson statistics and their P-values.

See Also

gof.sandwich for general distributions using Sandwich estimation of covariance function; gof for generic functions using imhof function.

Examples

x0=runif(n=100,min=-1,max=1)
gof.uniform.bootstrap(x0,M=100)

x1=rnorm(n=100,mean=0,sd=1)
gof.normal.bootstrap(x1,M=100)

x2=rgamma(n=100,shape=1,scale=1)
gof.gamma.bootstrap(x2,M=100)

x3=rlogis(n=100,location=0,scale=1)
gof.logistic.bootstrap(x3,M=100)

x4= rmutil::rlaplace(n=100,m=0,s=1)
gof.laplace.bootstrap(x4,M=100)

x5=rweibull(n=100,shape=1,scale=1)
gof.weibull.bootstrap(x5,M=100)
x5_log=log(x5)
gof.extremevalue.bootstrap(x5_log,M=100)

x6=rexp(n=100,rate=1/2)
gof.exp.bootstrap(x6,M=100)

EDFtest documentation built on Oct. 25, 2021, 9:10 a.m.