A unified approach in addition or deletion of two level factorial designs

H. Evangelaras, C. Koukouvinos and P. Mantas

Statistics & Probability Letters, 2002, vol. 59, issue 2, 125-133

Abstract: Suppose it is desired to have an optimal resolution III fraction of a 2p factorial in n runs where n[reverse not equivalent]1 (mod 4) or n[reverse not equivalent]3 (mod 4). If n[reverse not equivalent]1 (mod 4), we have to decide if we should add a run in a nxp submatrix of a Hadamard matrix of order n, say Hn or, alternatively, if we should delete three runs from a (n+4)xp submatrix of a Hadamard matrix of order n+4, say Hn+4, in an optimal manner, respectively. Similarly, when n[reverse not equivalent]3 (mod 4), we have to decide between optimally adding three more runs to a nxp submatrix of Hn or optimally deleting a single run from a (n+4)xp submatrix of Hn+4. The question to be studied is whether both strategies give designs that are equally efficient in terms of a well defined optimality criterion. We show that, in both cases, for p=3 both strategies give equally efficient designs under the D- or the A-optimality criterion. When n[reverse not equivalent]1 (mod 4) and p>3, both criteria show that the "addition" design is always better than the "deletion" design. However, when n[reverse not equivalent]3 (mod 4) and p>3, the choice of the most efficient design varies as p enlarges.

Keywords: Resolution; III; design; Fractional; factorial; D-optimality; A-optimality; Hadamard; matrix (search for similar items in EconPapers)
Date: 2002
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0167-7152(02)00111-6
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:stapro:v:59:y:2002:i:2:p:125-133

Ordering information: This journal article can be ordered from
http://www.elsevier.com/wps/find/supportfaq.cws_home/regional
https://shop.elsevie ... _01_ooc_1&version=01

Access Statistics for this article

Statistics & Probability Letters is currently edited by Somnath Datta and Hira L. Koul

More articles in Statistics & Probability Letters from Elsevier
Bibliographic data for series maintained by Catherine Liu ().