Comparer des méthodes

	Analyse conjointe bayésienne	Modélisation bayésienne de mélanges
Domaine	Statistique	Statistique
Famille	Latent structure	Latent structure
Année d'origine	1995	1997 (Richardson & Green Bayesian formulation)
Auteur d'origine	Allenby & Ginter (hierarchical Bayes formulation); conjoint roots in Luce & Tukey (1964)	Richardson & Green (seminal Bayesian treatment, 1997); broader Bayesian mixture roots trace to Dempster, Laird & Rubin (EM, 1977) and Titterington, Smith & Makov (1985)
Type	Preference measurement / Bayesian hierarchical model	Latent-class / model-based clustering
Source fondatrice	Allenby, G. M. & Ginter, J. L. (1995). Using extremes to design products and segment markets. Journal of Marketing Research, 32(4), 392–403.	Fruhwirth-Schnatter, S., Celeux, G. & Robert, C. P. (Eds.) (2019). Handbook of Mixture Analysis. CRC Press / Chapman & Hall. ISBN: 9780367733995
Alias	Bayesian CA, hierarchical Bayes conjoint, HB conjoint, Bayesian preference modeling	Bayesian mixture model, BMM, Bayesian model-based clustering, Bayesian finite mixture
Apparentées	6	4
Résumé	Bayesian conjoint analysis estimates individual-level consumer preference weights for product attributes by combining conjoint choice tasks with a hierarchical Bayesian model. It yields part-worth utilities for each respondent rather than only group averages, enabling precise market simulation and segment discovery even from small per-person choice sets.	Bayesian mixture modeling represents the population as a weighted sum of K component distributions and estimates all unknowns — mixing weights, component parameters, and even the number of components — through posterior inference. It extends classical mixture analysis by placing priors on every parameter and quantifying uncertainty over latent group assignments rather than treating them as fixed.