aequationes mathematicae

, Volume 32, Issue 1, pp 74–86 | Cite as

Further evidence against independence preservation in expert judgement synthesis

  • Christian Genest
  • Carl G. Wagner
Research Papers


When a decision maker chooses to form his/her own probability distribution by combining the opinions of a number of experts, it is sometimes recommended that he/she should do so in such a way as to preserve any form of expert agreement regarding the independence of the events of interest. In this paper, we argue against this recommendation. We show that for those probability spaces which contain at least five points, a large class of seemingly reasonable combination methods excludes all independence preserving formulas except those which pick a single expert. In the case where at most four alternatives are present, the same conditions admit a richer variety of non-dictatorial methods which we also characterize. In the discussion, we give our reasons for rejecting independence preservation in expert judgement synthesis.

AMS (1980) subject classification

Primary 62A99 62C05 Secondary 39B40 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Abou-Zaid, S. H. S.,Functional equations and related measurements. M. Phil. thesis, University of Waterloo, Waterloo, Ontario, 1984.Google Scholar
  2. [2]
    Bordley, R. F.,A Bayesian model of group polarization. Org. Behav. Hum. Perform.32 (1983), 262–274.Google Scholar
  3. [3]
    French, S.,Updating of belief in the light of someone else's opinion. J. R. Statist. Soc. Ser. A143 (1980), 43–48.MATHMathSciNetGoogle Scholar
  4. [4]
    Genest, C.,A conflict between two axioms for combining subjective distributions. J. R. Statist. Soc. Ser. B46 (1984), 403–405.MATHMathSciNetGoogle Scholar
  5. [5]
    Genest, C.,Pooling operators with the marginalization property. Canad. J. Statist.12 (1984), 153–163.MATHMathSciNetGoogle Scholar
  6. [6]
    Genest, C., McConway, K. J. andSchervish, M. J.,Characterization of externally Bayesian pooling operators. Ann. Statist.14 (1986), 487–501.MathSciNetGoogle Scholar
  7. [7]
    Genest, C. andZidek, J. V.,Combining probability distributions: a critique and an annotated bibliography. Statist. Sci.1 (1986), 114–148.MathSciNetGoogle Scholar
  8. [8]
    Kannappan, Pl. andNg, C. T.,On functional equations and measures of information, II. J. Appl. Prob.17 (1980), 271–277.MathSciNetGoogle Scholar
  9. [9]
    Laddaga, R.,Lehrer and the consensus proposal. Synthese36 (1977), 473–477.CrossRefMATHGoogle Scholar
  10. [10]
    Lehrer, K. andWagner, C. G.,Probability amalgamation and the independence issue: a reply to Laddaga. Synthese55 (1983), 339–346.MathSciNetGoogle Scholar
  11. [11]
    Levi, I.,Consensus as shared agreement and outcome of inquiry. Synthese62 (1985), 3–11.CrossRefGoogle Scholar
  12. [12]
    Lindley, D. V. andSingpurwalla, N. D.,Reliability (and fault tree) analysis using expert opinions. J. Amer. Statist. Assoc.81 (1986), 87–90.MathSciNetGoogle Scholar
  13. [13]
    Madansky, A.,Externally Bayesian groups. Rand memo RM-4141-PR, The Rand Corporation, Santa Monica, CA, 1964.Google Scholar
  14. [14]
    McConway, K. J.,Marginalization and linear opinion pools. J. Amer. Statist. Assoc.76, (1981), 410–414.MATHMathSciNetGoogle Scholar
  15. [15]
    Ng, C. T.,Uniqueness theorems for a general class of functional equations. J. Austr. Math. Soc.11 (1970), 362–366.MATHGoogle Scholar
  16. [16]
    Raiffa, H.,Decision analysis: introductory lectures on choices under uncertainty. Addison-Wesley, Reading, MA, 1968.Google Scholar
  17. [17]
    Roberts, A. W. andVarberg, D. E.,Convex functions. Academic Press, New York, 1973.Google Scholar
  18. [18]
    Wagner, C. G.,Aggregating subjective probabilities: some limitative theorems. Notre Dame J. Formal Logic25 (1984), 233–240.CrossRefMATHMathSciNetGoogle Scholar
  19. [19]
    Winkler, R. L.,Combining probability distributions from dependent information sources. Manag. Sci.27 (1981), 479–488.MATHMathSciNetGoogle Scholar

Copyright information

© Birkhäuser Verlag 1987

Authors and Affiliations

  • Christian Genest
    • 1
    • 2
  • Carl G. Wagner
    • 1
    • 2
  1. 1.Department of Statistics and Actuarial ScienceUniversity of WaterlooWaterlooCanada
  2. 2.Department of MathematicsUniversity of TennesseeKnoxvilleUSA

Personalised recommendations