Application of Covariates Within Sawtooth Software's CBC/HB Program: Theory and Practical Example

Keywords

choice-based conjoint, hierarchical bayes, part-worth estimation, covariates

Abstract

Over the last decade, hierarchical Bayes (HB) estimation of part-worths has had a significant and positive impact on the analysis of discrete choice (CBC) data. Certainly, HB has been key to the emergence of CBC (Choice-Based Conjoint) as the most popular conjoint-related method (Sawtooth Software, 2003). Sawtooth Software released its CBC/HB software for estimating part-worths from CBC questionnaires in 1999, based on earlier work published by Greg Allenby and Peter Lenk (Allenby et al. 1995, Lenk et al. 1996), as well as workshops given by Allenby & Lenk at the American Marketing Association’s ART/Forum conferences. Although the focus of this paper is on the use of CBC/HB for analyzing CBC data, similar data such as MaxDiff will also benefit from the use of covariates.

Original Publication Citation

Orme, Bryan, John R. Howell (2009), "Application of Covariates within Sawtooth Software’s CBC/HB Program: Theory and Practical Example", Sawtooth Software Technical Papers,

Document Type

Conference Paper

Publication Date

2009

Publisher

Sawtooth Software Technical Papers

Language

English

College

Marriott School of Business

Department

Marketing

University Standing at Time of Publication

Assistant Professor

Link to Full Text

Share

COinS