Decision Analysis by Augmented Probability Simulation

We provide a generic Monte Carlo method to find the alternative of maximum expected utility in a decision analysis. We define an artificial distribution on the product space of alternatives and states, and show that the optimal alternative is the mode of the implied marginal distribution on the alternatives. After drawing a sample from the artificial distribution, we may use exploratory data analysis tools to approximately identify the optimal alternative. We illustrate our method for some important types of influence diagrams.

[1]  A. C. Miller,et al.  Discrete Approximations of Probability Distributions , 1983 .

[2]  Max Henrion,et al.  Propagating uncertainty in bayesian networks by probabilistic logic sampling , 1986, UAI.

[3]  Ross D. Shachter.  Evaluating Influence Diagrams , 1986, Oper. Res..

[4]  Judea Pearl,et al.  Fusion, Propagation, and Structuring in Belief Networks , 1986, Artif. Intell..

[5]  Judea Pearl,et al.  Evidential Reasoning Using Stochastic Simulation of Causal Models , 1987, Artif. Intell..

[6]  Adrian F. M. Smith,et al.  Bayesian computation via the gibbs sampler and related markov chain monte carlo methods (with discus , 1993 .

[7]  David J. Spiegelhalter,et al.  Local computations with probabilities on graphical structures and their application to expert systems , 1990 .

[8]  Ross D. Shachter.  Probabilistic Inference and Influence Diagrams , 1988, Oper. Res..

[9]  Ross D. Shachter,et al.  Simulation Approaches to General Probabilistic Inference on Belief Networks , 2013, UAI.

[10]  C. Robert Kenley,et al.  Gaussian influence diagrams , 1989 .

[11]  Adrian F. M. Smith,et al.  Sampling-Based Approaches to Calculating Marginal Densities , 1990 .

[12]  Tomas Hrycej,et al.  Gibbs Sampling in Bayesian Networks , 1990, Artif. Intell..

[13]  Andrew L. Rukhin,et al.  Tools for statistical inference , 1991 .

[14]  Ross D. Shachter,et al.  Decision Making Using Probabilistic Inference Methods , 1992, UAI.

[15]  Jeremy York,et al.  Use of the Gibbs Sampler in Expert Systems , 1992, Artif. Intell..

[16]  James E. Smith.  Moment Methods for Decision Analysis , 1993 .

[17]  Prakash P. Shenoy,et al.  A comparison of graphical techniques for decision analysis , 1994 .

[18]  Adrian F. M. Smith,et al.  Simple conditions for the convergence of the Gibbs sampler and Metropolis-Hastings algorithms , 1994 .

[19]  L. Tierney.  Markov Chains for Exploring Posterior Distributions , 1994 .

[20]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo conver-gence diagnostics: a comparative review , 1996 .

[21]  C. Robert.  Convergence Control Methods for Markov Chain Monte Carlo Algorithms , 1995 .

[22]  Bruce Abramson,et al.  Decision analytic networks in artificial intelligence , 1995 .

[23]  Ali Jenzarli,et al.  Solving Influence Diagrams Using Gibbs Sampling , 1995, AISTATS.

[24]  Peter Müller,et al.  Optimal Design via Curve Fitting of Monte Carlo Experiments , 1995 .

[25]  Malcolm Pradhan,et al.  Optimal Monte Carlo Estimation of Belief Network Inference , 1996, UAI.

[26]  Mark J. Brewer,et al.  A comparison of hybrid strategies for Gibbs sampling in mixed graphical models , 1996 .

[27]  Adrian F. M. Smith,et al.  Bayesian Statistics 5. , 1998 .

[28]  A E Gelfand,et al.  Approaches for optimal sequential decision analysis in clinical trials. , 1998, Biometrics.

[29]  J L Palmer,et al.  Bayesian optimal design in population models for haematologic data. , 1998, Statistics in medicine.

[30]  Concha Bielza,et al.  A Comparison of Graphical Techniques for Asymmetric Decision Problems , 1999 .

[31]  Francisco J. Samaniego,et al.  The Practice of Bayesian Analysis , 1999, Technometrics.

[32]  Prakash P. Shenoy,et al.  A Forward Monte Carlo Method For Solving Influence Diagrams Using Local Computation , 2000 .