Summary
Anexpert (for You) is here defined as someone who shares Your world-view, but knows more than You do, so that were She to reveal Her current opinion to You, You would adopt it as Your own. When You have access to different experts, with differing information, You require acombination formula to aggregate their various opinions. A number of formulae have been suggested, but here we explore the fundamental requirement ofcoherence to relate such a formula to Your joint distribution for the experts' opinions. In particular, in the context of opinions about an uncertain eventA, we investigate coherence properties of the linear, harmonic and logarithmic opinion pools. Some general results on coherence of the joint forecast distribution are also developed.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
Berger, J. O. and Mortera, J. (1991). Bayesian limited communication.J.Statist. Planning and Inference 28, 1–24.
Bordley, R. F. (1982). A multiplicative formula for aggregating probability assessments.Manag. Sci. 28, 1137–1148.
Carnap, R. (1950).Logical Foundations of Probability. Chicago: University Press.
Clemen, R. T. (1985). Extraneous expert information.J. Forecasting 4, 329–348.
Clemen, R. T. (1986). Calibration and the aggregation of probabilities.Manag. Sci. 32, 312–314.
Clemen, R. T. and Murphy, A. H. (1986). Objective and subjective precipitation probability forecasts: statistical analysis of some interrelationships.Weather and Forecasting 1, 56–65.
Dawid, A. P. (1979). Conditional independence in statistical theory.J. Roy. Statist. Soc. B 41, 1–31, (with discussion).
Dawid, A. P. (1982). The well-calibrated Bayesian.J. Amer. Statist. Assoc. 77, 605–610.
Dawid, A. P. (1986). Probability forecasting.Encyclopedia of the Statistical Sciences (S. Kotz, N. L. Johnson and C. B. Read, eds.), vol.7. New York: Wiley, 210–218.
DeGroot, M. H. (1974). Reaching a consensus.J. Amer. Statist. Assoc. 69, 118–121.
DeGroot, M. H. (1988). A Bayesian view of assessing uncertainty and comparing expert opinion.J. Statist. Planning and Inference 20, 295–306.
DeGroot, M. H. and Eriksson, E. A. (1985). Probability forecasting, stochastic dominance and the Lorenz curve.Bayesian Statistics 2, (J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.), Amsterdam: North-Holland, 99–118, (with discussion).
DeGroot M. H. and Fienberg, S. E. (1983). The comparison and evaluation of forecasters.The Statistician 32, 12–22.
DeGroot, M. H. and Mortera, J. (1991). Optimal linear opinion pools.Manag. Sci. 37, 546–558.
French, S. (1980). Updating belief in the light of someone else's opinion.J. Roy. Statist. Soc. A 143, 43–48.
French, S. (1981). Consensus of opinion.Europ. J. Oper. Res. 7, 332–340.
French, S. (1985). Group consensus probability distribution: a critical survey.Bayesian Statistics 2 (J. M. Bernardo M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.), Amsterdam: North-Holland, 183–201, (with discussion).
French, S. (1986). Calibration and the expert problem.Manag. Sci. 32, 315–320.
Genest, C. (1984). A characterization theorem for externally bayesian groups.Ann. Statist. 12, 1100–1105.
Genest, C. and McConway K. J. (1990). Allocating the weights in the linear opinion pool.J. Forecasting 9, 53–73.
Genest, C., McConway K. J. and Schervish, M. J. (1986). Characterization of externally Bayesian pooling operators.Ann. Statist. 14, 487–501.
Genest, C. and Schervish, M. J. (1985). Modelling expert judgments for Bayesian updating.Ann. Statist. 13, 1198–1212.
Genest, C. and Zidek, J. V. (1986). Combining probability distributions: a critique and an annotated bibliography.Statist. Sci. 1, 114–148.
Gutmann, S., Kemperman, J. H. B., Reeds J. A. and Shepp, L. A. (1991). Existence of probability measures with given marginals.Annals of Prob. 19, 1781–1797.
Jeffreys H. (1939).Theory of Probability. Oxford: Clarendon Press.
Kellerer, H. G. (1961). Funktionen und Produkträumen mit vorgegeben Marginal-Funktionen.Math. Ann. 144, 323–344.
Keynes, J. M. (1921).A Treatise on Probability. London. Macmillan.
Lauritzen, S. L. (1980).Statistical Models as Extremal Families, Aalborg: University Press.
Lindley, D. V. (1982). The improvement of probability judgementsJ. Roy. Statist. Soc. B 145, 117–126.
Lindley, D. V. (1985). Reconciliation, of discrete probability distributions.Bayesian Statistics 2 (J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.), Amsterdam: North-Holland, 375–390, (with discussion).
Lindley, D. V. (1986). Another look at an axiomatic approach to expert resolution.Manag. Sci. 32, 303–306.
McConway, K. J. (1981). Marginalization and linear opinion pools.J. Amer. Statist. Assoc. 76, 410–414.
Madansky, A. (1964). Externally Bayesian groups. RAND Memo RM-4141-PR. Santa Monica, CA: The Rand Corporation.
Morris, P. A. (1983). An axiomatic approach to expert resolution.Manag. Sci. 29, 24–32.
Murphy, A. H. and Winkler R. L. (1992). Diagnostic verification of probability forecasts.Int. J. Forecasting 7, 435–455.
Schervish, M. J. (1986). Comments on some axioms for combining expert judgments.Manag. Sci. 32, 306–312.
Singpurwalla, N. D. (1988). Discussion of “Gaining weight: A Bayesian Approach” by Bayarri, M. J. and DeGroot, M. H.Bayesian Statistics 2 (J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.), Amsterdam: North-Holland, 39–42.
Stone, M. (1961). The opinion pool.Ann. Math. Statist. 32, 1339–1342.
Strassen, V. (1965). The existence of probability measures with given marginals.Ann. Math. Statist. 36, 423–439.
Winkler, R. L. (1981). Combining probability distributions from dependent information sources.Manag. Sci. 27, 479–488.
Winkler, R. L. (1986). Expert resolution.Manag. Sci. 32, 298–303.
Additional references in the discussion
Arrow, K. J. (1951).Social Choice and Individual Values. New York: Wiley.
Clemen, R. T. (1987). Combining overlapping information.Manag. Sci. 33, 373–380.
Clemen, R. T. and Winkler, R. L. (1990). Unanimity and compromise among probability forecasters.Manag. Sci. 36, 767–779.
Clemen, R. T. and Jouini, M. N. (1996). Copula models for aggregating expert opinions.Operations Research 44, (to appear).
Cooke, R. M. (1993). The ill advised BayesianISBA Newsletter 2, 2–3.
Dalkey N. C. (1972). An impossibility theorem for group probability functionsTech. Rep. P-4862. Santa Monica, CA: RAND Corporation.
Draper, D. (1995). Assessment and propagation of model uncertainty.J. Roy. Statist. Soc. B 57, 45–97, (with discussion).
French, S. (1994). Utility: probability's younger twin.Aspects of Uncertainty: a Tribute to D. V. Lindley (P. R. Freeman, and A. F. M. Smith, eds.), Chichester: Wiley, 171–180.
Harper, F. T.; Goossens, L. H. J.; Cooke, R. M.; Helton, J. C.; Hora, S. C.; Jones, J. A.; Kraan, B. C. P.; Lui, C.; McKay, M. D.; Miller, L. A.; Päsler Sauer, J. and Young, M. L. (1994).Joint USNRC/CEC Consequence Uncertainty Study: Summary of Objectives, Approach, Application and Results for the Dispersion and Deposition Uncertainty Assessments. NUREG/CR-6244, EUR 15855, SAND94-1453. Sandia National Laboratories and Delft University of Technology.
Kelly, F. S. (1978).Arrow Impossibility Theorems. New York: Academic Press.
Makridakis, S. and Winkler, R. L. (1983). Average of forecasts: some empirical results.Manag. Sci. 29, 987–996.
Author information
Authors and Affiliations
Additional information
Morrie DeGroot passed away on 2 November, 1989. This work was initiated by Dawid and DeGroot and was completed by Dawid and Mortera.
Rights and permissions
About this article
Cite this article
Dawid, A.P., DeGroot, M.H., Mortera, J. et al. Coherent combination of experts' opinions. Test 4, 263–313 (1995). https://doi.org/10.1007/BF02562628
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02562628