Zero probabilities in stochastic independence
暂无分享,去创建一个
In a previous paper we pointed out the lack of the classic definition of stochastic independence: a stronger one (that avoids the main inconsistencies related to events of probability 0 or 1) has been introduced in that paper. We give here a new “improved” definition of independence of two events (coinciding with the former one when some relevant events have probabilities different from 0 and 1): it allows to deepen and clarify further aspects (in particular those connected with the possible “symmetry” of this important probabilistic concept) and to establish a characterization theorem.
[1] Giulianella Coletti,et al. Coherent Numerical and Ordinal Probabilistic Assessments , 1994, IEEE Trans. Syst. Man Cybern. Syst..
[2] Romano Scozzafava,et al. Characterization of coherent Conditional Probabilities as a Tool for their Assessment and Extension , 1996, Int. J. Uncertain. Fuzziness Knowl. Based Syst..
[3] Giulianella Coletti,et al. Conditioning and inference in intelligent systems , 1999, Soft Comput..
[4] Giulianella Coletti,et al. Null events and stochastical independence , 1998, Kybernetika.