Power assessment for tests of the equality of several proportions
Power assessment for tests of the equality of several proportions
In this paper the elementary statistical problem of testing the equality of several Bernoulli probabilities is considered, and attention is specifically directed to forming an assessment of the power properties of such a test. In particular, power levels are considered under a simple restriction on the range of the Bernoulli probabilities. A test procedure based on the range of the arcsin-root transformations of the observed proportions is considered, and it is shown how power levels may be calculated both exactly and under asymptotic assumptions.
Bernoulli data, power, least favourable configuration, sample size determination
19-30
Hayter, A.J.
55bd07a5-db1d-4d3d-8c87-b307485420d9
Liu, Wei
b64150aa-d935-4209-804d-24c1b97e024a
1990
Hayter, A.J.
55bd07a5-db1d-4d3d-8c87-b307485420d9
Liu, Wei
b64150aa-d935-4209-804d-24c1b97e024a
Hayter, A.J. and Liu, Wei
(1990)
Power assessment for tests of the equality of several proportions.
Communications in Statistics: Theory and Methods, 19 (1), .
(doi:10.1080/03610929008830184).
Abstract
In this paper the elementary statistical problem of testing the equality of several Bernoulli probabilities is considered, and attention is specifically directed to forming an assessment of the power properties of such a test. In particular, power levels are considered under a simple restriction on the range of the Bernoulli probabilities. A test procedure based on the range of the arcsin-root transformations of the observed proportions is considered, and it is shown how power levels may be calculated both exactly and under asymptotic assumptions.
This record has no associated files available for download.
More information
Published date: 1990
Keywords:
Bernoulli data, power, least favourable configuration, sample size determination
Organisations:
Statistical Sciences Research Institute
Identifiers
Local EPrints ID: 381076
URI: http://eprints.soton.ac.uk/id/eprint/381076
ISSN: 0361-0926
PURE UUID: 21991ed3-67b7-4cfc-93d9-bde8843f53bb
Catalogue record
Date deposited: 05 Oct 2015 11:30
Last modified: 15 Mar 2024 02:43
Export record
Altmetrics
Contributors
Author:
A.J. Hayter
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
