Efficient paired choice designs with fewer choice pairs

For paired choice experiments, two new construction methods of designs are proposed for the estimation of the main effects. In many cases, these designs require about 30–50% fewer choice pairs than the existing designs and at the same time have reasonably high D -efficiencies for the estimation of t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Metrika 2017-04, Vol.80 (3), p.309-317
Hauptverfasser: Dey, Aloke, Singh, Rakhi, Das, Ashish
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:For paired choice experiments, two new construction methods of designs are proposed for the estimation of the main effects. In many cases, these designs require about 30–50% fewer choice pairs than the existing designs and at the same time have reasonably high D -efficiencies for the estimation of the main effects. Furthermore, as against the existing efficient designs, our designs have higher D -efficiencies for the same number of choice pairs.
ISSN:0026-1335
1435-926X
DOI:10.1007/s00184-016-0605-9