Graphical Models for Groups: Belief Aggregation and Risk Sharing

We investigate the practical value of using graphical models to aid in two fundamental problems of group coordination: (1) belief aggregation and (2) risk sharing. We identify restrictive conditions under which graphical models can be useful in both settings. We show that the output of the logarithmic opinion pool (LogOP) can be represented as a Markov network (MN) or a decomposable Bayesian network (BN), and give an algorithm for doing so. We show that a securities market structured like a decomposable BN can support optimal risk sharing, if all agents have exponential utility and all of their Markov independencies coincide with the market structure. On the other hand, most of our results are negative, taking the form of impossibility theorems. We show that no belief aggregation function can maintain all independencies representable in a BN. Neither can an aggregation computation be decomposed into local computations on graph subsets. We show that computing query outputs of LogOP or the linear opinion pool (LinOP) is NP-hard. Except in fairly restrictive settings, structuring securities markets according to unanimously agreed upon independencies may be of no help in supporting optimal risk sharing because agents' behavioral independencies change as they engage in securities trade.

[1]  C. Genest,et al.  Further evidence against independence preservation in expert judgement synthesis , 1987 .

[2]  R. Nau.  Coherent decision analysis with inseparable probabilities and utilities , 1995 .

[3]  Norman C. Dalkey,et al.  Toward a theory of group estimation , 1975 .

[4]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems , 1988 .

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

[6]  J. Crisp,et al.  The Delphi method? , 1997, Nursing research.

[7]  B. M. Hill,et al.  Theory of Probability , 1990 .

[8]  Robert L. Winkler,et al.  Expert resolution , 1986 .

[9]  Murray Turoff,et al.  The Delphi Method: Techniques and Applications , 1976 .

[10]  Ross D. Shachter,et al.  Directed reduction algorithms and decomposable graphs , 1990, UAI.

[11]  M. Degroot.  Reaching a Consensus , 1974 .

[12]  R. L. Winkler.  Combining Probability Distributions from Dependent Information Sources , 1981 .

[13]  Peter A. Morris,et al.  Combining Expert Judgments: A Bayesian Approach , 1977 .

[14]  K. Arrow.  The Role of Securities in the Optimal Allocation of Risk-bearing , 1964 .

[15]  Howard E. Reinhardt.  Theory of Probability: A Critical Introductory Treatment, Vol. 2 (Bruno de Finetti) , 1978 .

[16]  E. Rosenblueth,et al.  COMBINATION OF EXPERT OPINIONS , 1992 .

[17]  Robert Laddaga,et al.  Lehrer and the consensus proposal , 1977, Synthese.

[18]  M. Degroot,et al.  Optimal linear opinion pools , 1991 .

[19]  A. Mas-Colell,et al.  Microeconomic Theory , 1995 .

[20]  R. L. Winkler,et al.  Separating probability elicitation from utilities , 1988 .

[21]  T. Speed,et al.  Markov Fields and Log-Linear Interaction Models for Contingency Tables , 1980 .

[22]  Chiuh-Cheng Chyu.  Decomposable Probabilistic Influence Diagrams , 1989 .

[23]  Peter A. Morris,et al.  Decision Analysis Expert Use , 1974 .

[24]  Keung-Chi Ng,et al.  Probabilistic multi-knowledge-base systems , 1994, Applied Intelligence.

[25]  Michael P. Wellman,et al.  Compact Securities Markets for Pareto Optimal Reallocation of Risk , 2000, UAI.

[26]  Álvaro E. Faria,et al.  Conditional external Bayesianity in decomposable influence diagrams , 1996 .

[27]  Ton Kloks.  Treewidth, Computations and Approximations , 1994, Lecture Notes in Computer Science.

[28]  Christian Genest,et al.  Combining Probability Distributions: A Critique and an Annotated Bibliography , 1986 .

[29]  Robert A. Jacobs,et al.  Methods For Combining Experts' Probability Assessments , 1995, Neural Computation.

[30]  Howard Raiffa,et al.  Decision analysis: introductory lectures on choices under uncertainty. 1968. , 1969, M.D.Computing.

[31]  C. Genest.  Pooling operators with the marginalization property , 1984 .

[32]  Carl G. Wagner.  Aggregating subjective probabilities: some limitative theorems , 1984, Notre Dame J. Formal Log..

[33]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[34]  C. Genest.  A Characterization Theorem for Externally Bayesian Groups , 1984 .

[35]  Yang Xiang,et al.  A Probabilistic Framework for Cooperative Multi-Agent Distributed Interpretation and Optimization of Communication , 1996, Artif. Intell..

[36]  Bruce Abramson,et al.  Some Complexity Considerations in the Combination of Belief Networks , 1993, UAI.

[37]  Keith Lehrer,et al.  Probability amalgamation and the independence issue: A reply to Laddaga , 1983, Synthese.

[38]  Jacques H. Dreze,et al.  Market allocation under uncertainty , 1970 .

[39]  H. Varian.  The Arbitrage Principle in Financial Economics , 1987 .

[40]  A. H. Murphy,et al.  “Good” Probability Assessors , 1968 .

[41]  E. Rowland.  Theory of Games and Economic Behavior , 1946, Nature.

[42]  Michael P. Wellman,et al.  Representing Aggregate Belief through the Competitive Equilibrium of a Securities Market , 1997, UAI.

[43]  Società italiana di fisica,et al.  Accelerated Life Testing and Experts' Opinion in Reliability , 1990 .

[44]  M. Schervish,et al.  Characterization of Externally Bayesian Pooling Operators , 1986 .

[45]  Bruce Abramson,et al.  The Topological Fusion of Bayes Nets , 1992, UAI.

[46]  Michael P. Wellman,et al.  Graphical Representations of Consensus Belief , 1999, UAI.

[47]  G. Yule.  NOTES ON THE THEORY OF ASSOCIATION OF ATTRIBUTES IN STATISTICS , 1903 .

[48]  C. Genest.  A Conflict between Two Axioms for Combining Subjective Distributions , 1984 .

[49]  Robert L. Winkler,et al.  The Consensus of Subjective Probability Distributions , 1968 .

[50]  Keung-Chi Ng,et al.  Consensus diagnosis: a simulation study , 1992, IEEE Trans. Syst. Man Cybern..

[51]  Richard E. Neapolitan,et al.  Probabilistic reasoning in expert systems - theory and algorithms , 2012 .

[52]  Robert L. Winkler,et al.  Aggregating Point Estimates: A Flexible Modeling Approach , 1993 .

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

[54]  Jon Atli Benediktsson,et al.  Consensus theoretic classification methods , 1992, IEEE Trans. Syst. Man Cybern..

[55]  K. McCardle,et al.  Arbitrage, rationality, and equilibrium , 1991 .

[56]  M. West,et al.  Modelling Probabilistic Agent Opinion , 1992 .

[57]  I. J. Myung,et al.  Maximum Entropy Aggregation of Expert Predictions , 1996 .

[58]  William B. Levy,et al.  Maximum entropy aggregation of individual opinions , 1994, IEEE Trans. Syst. Man Cybern..

[59]  David Haussler,et al.  How to use expert advice , 1993, STOC.

[60]  Judea Pearl,et al.  From Bayesian networks to causal networks , 1995 .

[61]  Michael P. Wellman,et al.  A Market Framework for Pooling Opinions , 2001 .

[62]  Francisco J. Samaniego.  Accelerated Life Testing and Experts' Opinions in Reliability , 1990 .

[63]  Gregory F. Cooper,et al.  The Computational Complexity of Probabilistic Inference Using Bayesian Belief Networks , 1990, Artif. Intell..

[64]  R. Cooke.  Experts in Uncertainty: Opinion and Subjective Probability in Science , 1991 .

[65]  D. Geiger.  Graphoids: a qualitative framework for probabilistic inference , 1990 .

[66]  Anna Nagurney,et al.  Foundations of Financial Economics , 1997 .