eprintid: 2194
rev_number: 8
eprint_status: archive
userid: 6
dir: disk0/00/00/21/94
datestamp: 2019-03-07 13:42:43
lastmod: 2019-03-07 13:42:43
status_changed: 2019-03-07 13:42:43
type: article
metadata_visibility: show
creators_name: Marsaglia, George
creators_name: Marsaglia, John
corp_creators: The Florida State University
corp_creators: Western Oregon University
title: Evaluating the Anderson-Darling Distribution
subjects: MP2_2
divisions: SHEER
full_text_status: none
abstract: Except for n=1, only the limit as n→∞ for the distribution of the Anderson-Darling test for uniformity
has been found, and that in so complicated a form that published values for a few percentiles had to be
determined by numerical integration, saddlepoint or other approximation methods. We give here our method
for evaluating that asymptotic distribution to great accuracy—directly, via series with two-term recursions.
We also give, for any particular n, a procedure for evaluating the distribution to the fourth digit, based on
empirical CDF’s from samples of size 1010
.
date: 2004
date_type: published
publication: Journal of Statistical Software
volume: 9
number: 2
publisher: Foundation for Open Access Statistics
id_number: doi:10.18637/jss.v009.i02
issn: 1548-7660
official_url: http://doi.org/10.18637/jss.v009.i02
access_IS-EPOS: limited
software_references: AD_Test_for_Magnitude_Distribution
owner: Publisher
citation:   Marsaglia, George and Marsaglia, John  (2004) Evaluating the Anderson-Darling Distribution.  Journal of Statistical Software, 9 (2).  DOI: https://doi.org/10.18637/jss.v009.i02 <https://doi.org/10.18637/jss.v009.i02>