Next: Independence
Up: QUASI-BAYESIAN THEORY
Previous: Lower and upper values
Convex sets of conditional
distributions are used to represent conditional beliefs. Inference
is performed by applying Bayes rule to each distribution in a
jont credal set. The posterior credal set is the union of all
posterior distributions obtained in this process, and the
vertices of the posterior credal set are obtained by applying
Bayes rule to all vertices of the joint credal set [28].
Denote by K(X|Y) the collection of credal sets K(X|A) for all
events A defined by a variable Y.
Fabio Gagliardi Cozman
1998-07-03