scientific article; zbMATH DE number 4115795

From MaRDI portal
Publication:4730655





[For the entire collection see Zbl 0676.00016.] Influence diagrams were developed as a Bayesian computer-aided modeling tool by \textit{R. A. Howard} and \textit{J. E. Matheson} [Influence diagrams, in Readings in Principles and Applications of Decision Analysis (1984)]. Influence diagrams also provide a graphical representation of conditional independence for random quantities. An algorithm for solving Bayesian decision analysis problems through infuence diagram manipulations was constructed by \textit{R. D. Shachter} [Evaluating influence diagrams. A. P. Basu (ed.), Reliability and Quality control (1986), p. 321]. This algorithm is useful in statistical applications, especially in experimental design. The influence diagram is also a generalization of the fault tree representation used in engineering risk analysis. The influence diagram is a modeling alternative to the decision tree. Since the order of event expansion required by the decision tree is rarely the natural order in which to assess the decision maker's information, the influence diagram may be more useful. This lecture is largely tutorial since the basic ideas were developed elsewhere. However, some of the proofs are original. We begin the discussion with influence diagrams having only probabilistic nodes. Deterministic and decision nodes are treated in later sections.




Cited in
(29)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4730655)