Combining the advantages of one-sided and two-sided test procedures for comparing several treatments with a control.
Combining the advantages of one-sided and two-sided test procedures for comparing several treatments with a control.
Consider the standard problem of investigating whether any of k treatments are better than a control. A two-sided procedure provides both upper and lower bounds on the differences between each treatment and the control, whereas a one-sided procedure only provides lower bounds on these differences. However, the one-sided procedure allows sharper inferences regarding which treatments can be declared to be better than the control. In this paper we develop a new procedure which combines the good aspects of both the one-sided and the two-sided procedures. This new procedure maintains the inferential sensitivity of the one-sided procedure while also providing both upper and lower bounds on the differences between each treatment and the control. A new set of critical points is needed which are tabulated for the balanced case where the treatment sample sizes are all equal. More generally, the new procedure is applicable to any set of correlated or uncorrelated parameter estimates.
acceptance sets, comparisons with a control, critical points, directional decisions, orthogonal contrasts, simultaneous confidence intervals
81-99
Hayter, A.J.
55bd07a5-db1d-4d3d-8c87-b307485420d9
Miwa, Tetsuhisa
cb4e7293-e076-437a-b31c-b5dd6c6a3b4a
Liu, Wei
b64150aa-d935-4209-804d-24c1b97e024a
2000
Hayter, A.J.
55bd07a5-db1d-4d3d-8c87-b307485420d9
Miwa, Tetsuhisa
cb4e7293-e076-437a-b31c-b5dd6c6a3b4a
Liu, Wei
b64150aa-d935-4209-804d-24c1b97e024a
Hayter, A.J., Miwa, Tetsuhisa and Liu, Wei
(2000)
Combining the advantages of one-sided and two-sided test procedures for comparing several treatments with a control.
Journal of Statistical Planning and Inference, 86 (1), .
(doi:10.1016/S0378-3758(99)00161-5).
Abstract
Consider the standard problem of investigating whether any of k treatments are better than a control. A two-sided procedure provides both upper and lower bounds on the differences between each treatment and the control, whereas a one-sided procedure only provides lower bounds on these differences. However, the one-sided procedure allows sharper inferences regarding which treatments can be declared to be better than the control. In this paper we develop a new procedure which combines the good aspects of both the one-sided and the two-sided procedures. This new procedure maintains the inferential sensitivity of the one-sided procedure while also providing both upper and lower bounds on the differences between each treatment and the control. A new set of critical points is needed which are tabulated for the balanced case where the treatment sample sizes are all equal. More generally, the new procedure is applicable to any set of correlated or uncorrelated parameter estimates.
This record has no associated files available for download.
More information
Published date: 2000
Keywords:
acceptance sets, comparisons with a control, critical points, directional decisions, orthogonal contrasts, simultaneous confidence intervals
Organisations:
Statistics
Identifiers
Local EPrints ID: 30098
URI: http://eprints.soton.ac.uk/id/eprint/30098
ISSN: 0378-3758
PURE UUID: f9c0904f-1353-479e-a078-28d682db7954
Catalogue record
Date deposited: 20 Jul 2006
Last modified: 16 Mar 2024 02:42
Export record
Altmetrics
Contributors
Author:
A.J. Hayter
Author:
Tetsuhisa Miwa
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
