KIT | KIT-Bibliothek | Impressum | Datenschutz
Open Access Logo
§
Volltext
DOI: 10.5445/KSP/1000058747/05

TCA/HB Compared to CBC/HB for Predicting Choices Among Multi-Attributed Products

Baier, Daniel; Pełka, Marcin; Rybicka, Aneta; Schreiber, Stefanie

Abstract:
For some years, choice-based conjoint analysis (CBC) has demonstrated its superiority over other preference measurement alternatives. So, e.g., in a recent study on German and Polish cola consumers, the superiority of CBC over traditional conjoint analysis (TCA) was striking. As one reason for this superiority, the usage of hierarchical Bayes for CBC parameter estimation was mentioned (CBC/HB). This paper clarifies whether this really makes the difference: Hierarchical Bayes is also used for TCA parameter estimation (TCA/HB). The application to the above mentioned data shows, that this improves the predictive validity compared to TCA but is still inferior to CBC/HB in “high data quality cases". However, in “low data quality cases" TCA/HB is superior to CBC/HB.


Zugehörige Institution(en) am KIT Institut für Informationswirtschaft und Marketing (IISM)
Publikationstyp Zeitschriftenaufsatz
Jahr 2016
Sprache Englisch
Identifikator ISSN: 2363-9881
URN: urn:nbn:de:swb:90-677638
KITopen ID: 1000067763
Erschienen in Archives of Data Science, Series A
Band 1
Heft 1
Seiten 77-87
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page