ADGofTest (version 0.3)

ad.test: Anderson-Darling GoF test

Description

Implementation of the Anderson-Darling goodness of fit test.

Usage

ad.test(x, distr.fun, ...)

Arguments

x

a random sample from a possibly unknown continuous distribution

distr.fun

a named CDF, such as pnorm, punif, etc.

...

extra parameters for the distribution function above, such as location and scale parameters, etc.

Value

The output is an object of the class htest exactly like for the Kolmogorov-Smirnov test, ks.test. The statistic and p.value fields are the most relevant ones.

Details

If the distr.fun is provided, the function checks whether x is a iid sample from the distribution described by such CDF. Otherwise, whether they follow a uniform law.

References

G. and J. Marsaglia, "Evaluating the Anderson-Darling Distribution", Journal of Statistical Software, 2004

Examples

    set.seed( 123 )
    x <- runif( 100 )

    ad.test( x )$p.value

    ad.test( x, pnorm, 0, 1 )$p.value

    replicate( ad.test( rnorm( 100 ), pnorm )$p.value, 100 )