A comparison of design and model selection methods for supersaturated experiments


Marley, Christopher J. and Woods, David C. (2009) A comparison of design and model selection methods for supersaturated experiments. Southampton, UK, Southampton Statistical Sciences Research Institute, 22pp. (S3RI Methodology Working Papers, (M09/20) ). (Submitted)

Download

[img] PDF
Download (163Kb)

Description/Abstract

Various design and model selection methods are available for supersatu-
rated designs having more factors than runs but little research is available on
their comparison and evaluation. In this paper, simulated experiments are
used to evaluate the use of E(s2)-optimal and Bayesian D-optimal designs,
and to compare three analysis strategies representing regression, shrinkage
and a novel model-averaging procedure. Suggestions are made for choosing
the values of the tuning constants for each approach. Findings include that
(i) the preferred analysis is via shrinkage; (ii) designs with similar numbers
of runs and factors can be effective for a considerable number of active effects
of only moderate size; and (iii) unbalanced designs can perform well. Some
comments are made on the performance of the design and analysis methods
when effect sparsity does not hold

Item Type: Monograph (Working Paper)
Subjects: H Social Sciences > HA Statistics
Divisions: University Structure - Pre August 2011 > Southampton Statistical Sciences Research Institute
ePrint ID: 69458
Date Deposited: 17 Nov 2009
Last Modified: 27 Mar 2014 18:49
URI: http://eprints.soton.ac.uk/id/eprint/69458

Actions (login required)

View Item View Item

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