方法对比

	Hierarchical Bayes Choice Model	Nested Logit Brand Choice
领域	市场营销	市场营销
方法族	Regression model	Regression model
起源年份	2005	1978
提出者	Peter E. Rossi, Greg M. Allenby & Robert McCulloch	Daniel McFadden
类型	Hierarchical Bayesian random-coefficients discrete-choice model	Generalized-extreme-value discrete-choice model
开创性文献	Rossi, P. E., Allenby, G. M., & McCulloch, R. (2005). Bayesian Statistics and Marketing. John Wiley & Sons. ISBN: 9780470863671	McFadden, D. (1978). Modelling the Choice of Residential Location. In A. Karlqvist, L. Lundqvist, F. Snickars, & J. Weibull (Eds.), Spatial Interaction Theory and Planning Models (pp. 75-96). North-Holland. ISBN: 9780444851826
别名	HB Choice Model, Bayesian Random-Coefficients Logit, Hierarchical Bayesian Conjoint, Individual-Level Partworth Model	Nested Multinomial Logit, Hierarchical Choice Model, Tree-Structured Logit, GEV Nested Logit
相关	3	3
摘要	Hierarchical Bayes (HB) choice models estimate a separate set of preference weights — partworths — for every individual respondent, while borrowing strength across respondents through a shared population distribution. The model has two levels: at the lower level each person's choices follow a logit driven by their own coefficients, and at the upper level those individual coefficients are treated as draws from a common multivariate distribution whose mean and covariance are themselves estimated. Inference is Bayesian and proceeds by Markov chain Monte Carlo — typically Gibbs sampling with Metropolis steps — which yields a full posterior for each respondent's partworths rather than a single point estimate. The approach, codified by Rossi, Allenby, and McCulloch, solved a long-standing problem in choice modeling: how to recover genuine individual-level heterogeneity from the sparse data each person provides. Sparse individual estimates are stabilized by shrinkage toward the population mean, giving reliable person-level coefficients usable for segmentation, targeting, and realistic market simulation. HB is now the default estimator for conjoint and scanner-based choice analysis.	The nested logit model of brand choice relaxes the restrictive independence-of-irrelevant-alternatives (IIA) assumption of the standard multinomial logit by grouping similar alternatives into nests. Developed by Daniel McFadden as a member of the generalized-extreme-value (GEV) family, it allows the unobserved utilities of alternatives within the same nest to be correlated while keeping a tractable closed form. In a brand-choice setting the natural structure is a tree: consumers first effectively choose a category, sub-category, or product form and then a brand within it, with an inclusive-value term carrying the expected utility of the lower level up to the upper level. The dissimilarity parameter on each nest measures within-nest correlation and reduces to ordinary logit when it equals one. The result is a model whose substitution patterns are far more realistic than plain logit — a price cut on one brand draws disproportionately from its nest-mates — while remaining estimable by maximum likelihood. It is a workhorse for choice analysis when alternatives fall into obvious clusters.