Being, relating to, or denoting statistical methods based on → Bayes' theorem.
Referring to → Bayes' theorem.
Fr.: inférence bayésienne
An approach to → statistical analysis in which → unknowns to be estimated have a prior → probability distribution which combined with the information from data produces a posterior probability distribution for the target quantities.
Fr.: modèle bayésien
A mathematical framework described by the prior distribution of a random parameter and by the likelihood of the observations. In this framework, all information on the random parameter based on the observations is included in the posterior distribution which can be obtained using → Bayes' theorem (see, e.g., Andrieu et al., 2001, "An Introduction to Monte Carlo Methods for Bayesian Data Analysis," in Nonlinear Dynamics and Statistics, ed. A. I. Mees, Boston: Birkhäuser).
Bayesian model averaging (BMA)
miyângin-giri-ye Bayesi-e model
An approach to model selection in which one bases inference on an average of all possible models instead of a single best model. The BMA is largely used in various branches of knowledge to properly account for model uncertainty in performing predictions.