Comparer des méthodes

	Modèle de Markov bayésien	Analyse de Sensibilité Bayésienne
Domaine	Simulation	Simulation
Famille	Process / pipeline	Process / pipeline
Année d'origine	1990s–2000s	1984–1994
Auteur d'origine	Briggs, A.; Sculpher, M.; and broader Bayesian statistics community	Berger, J. O. (Bayesian robustness); Saltelli et al. (global SA integration)
Type	Probabilistic state-transition simulation	Uncertainty propagation and sensitivity quantification
Source fondatrice	Briggs, A., Sculpher, M., Claxton, K. (2006). Decision Modelling for Health Economic Evaluation. Oxford University Press, Oxford. ISBN: 9780198526629	Berger, J. O. (1994). An overview of robust Bayesian analysis. Test, 3(1), 5–124.
Alias	Bayesian Markov Chain Model, Bayesian State-Transition Model, BMM, Bayesian Cohort Simulation	BSA, Bayesian SA, Bayesian robustness analysis, prior sensitivity analysis
Apparentées	4	5
Résumé	A Bayesian Markov model is a state-transition simulation method that combines Markov chain cohort modeling with Bayesian statistical inference. By placing prior distributions on transition probabilities and updating them with observed data, the approach propagates full parameter uncertainty through the simulation, yielding posterior distributions over outcomes such as costs, life-years, or quality-adjusted life-years rather than single-point estimates.	Bayesian Sensitivity Analysis (BSA) combines Bayesian inference with sensitivity analysis to systematically quantify how uncertain model inputs — expressed as prior probability distributions — propagate through a model and influence outputs. It identifies which parameters most drive output variability, supporting robust conclusions under genuine uncertainty.