Robustness of a class of partial diallel cross designs to the unavailability of a complete block of observations.
Robustness of a class of partial diallel cross designs to the unavailability of a complete block of observations.
Complete and partial diallel cross designs are examined as to their construction and robustness against the loss of a block of observations. A simple generalized inverse is found for the information matrix of the line effects, which allows evaluation of expressions for the variances of the line-effect differences with and without the missing block. A-efficiencies, based on average variances of the elementary contrasts of the line-effects, suggest that these designs are fairly robust. The loss of efficiency is generally less than 10%, but it is shown that specific comparisons might suffer a loss of efficiency of as much as 40%.
partially balanced incomplete block designs, group divisible designs, information matrix, generalized inverse, eigenvalues, pairwise variances of treatment differences
145-160
Mansson, Ralph
9788f5c4-af4d-4cc5-94bc-a4922d637a5a
Prescott, Philip
cf0adfdd-989b-4f15-9e60-ef85eed817b2
2004
Mansson, Ralph
9788f5c4-af4d-4cc5-94bc-a4922d637a5a
Prescott, Philip
cf0adfdd-989b-4f15-9e60-ef85eed817b2
Mansson, Ralph and Prescott, Philip
(2004)
Robustness of a class of partial diallel cross designs to the unavailability of a complete block of observations.
Journal of Applied Statistics, 31 (2), .
(doi:10.1080/0266476032000148830).
Abstract
Complete and partial diallel cross designs are examined as to their construction and robustness against the loss of a block of observations. A simple generalized inverse is found for the information matrix of the line effects, which allows evaluation of expressions for the variances of the line-effect differences with and without the missing block. A-efficiencies, based on average variances of the elementary contrasts of the line-effects, suggest that these designs are fairly robust. The loss of efficiency is generally less than 10%, but it is shown that specific comparisons might suffer a loss of efficiency of as much as 40%.
This record has no associated files available for download.
More information
Published date: 2004
Keywords:
partially balanced incomplete block designs, group divisible designs, information matrix, generalized inverse, eigenvalues, pairwise variances of treatment differences
Organisations:
Statistics
Identifiers
Local EPrints ID: 29999
URI: http://eprints.soton.ac.uk/id/eprint/29999
ISSN: 0266-4763
PURE UUID: 32924731-70fb-4945-bff9-55f7c4a976c3
Catalogue record
Date deposited: 11 May 2006
Last modified: 15 Mar 2024 07:36
Export record
Altmetrics
Contributors
Author:
Ralph Mansson
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