Abstract
The proposed minimax procedure blends strict Bayesian methods with p values and confidence intervals or with default-prior methods. Two applications to hypothesis testing bring some implications to light. First, the blended probability that a point null hypothesis is true is equal to the p value or a lower bound of an unknown posterior probability, whichever is greater. As a result, the p value is reported instead of any posterior probability in the case of complete prior ignorance but is ignored in the case of a fully known prior. In the case of partial knowledge about the prior, the possible posterior probability that is closest to the p value is used for inference. The second application provides guidance on the choice of methods used for small numbers of tests as opposed to those appropriate for large numbers. Whereas statisticians tend to prefer a multiple comparison procedure that adjusts each p value for small numbers of tests, large numbers instead lead many to estimate the local false discovery rate (LFDR), a posterior probability of hypothesis truth. Each blended probability reduces to the LFDR estimate if it can be estimated with sufficient accuracy or to the adjusted p value otherwise.
Similar content being viewed by others
References
Alba R, Payton P, Fei Z, McQuinn R, Debbie P, Martin GB, Tanksley SD, Giovannoni JJ (2005) Transcriptome and selected metabolite analyses reveal multiple points of ethylene control during tomato fruit development. Plant Cell 17:2954–2965
Augustin T (2002) Expected utility within a generalized concept of probability—a comprehensive framework for decision making under ambiguity. Stat Pap 43(1):5–22
Augustin T (2004) Optimal decisions under complex uncertainty—basic notions and a general algorithm for data-based decision making with partial prior knowledge described by interval probability. Zeitschrift fur Angewandte Mathematik und Mechanik 84(10–11):678–687
Balch MS (2012) Mathematical foundations for a theory of confidence structures. Int J Approx Reason 53(7):1003–1019
Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57:289–300
Berger JO (1984) Robustness of Bayesian analyses. Studies in Bayesian econometrics. North-Holland, Ch. The robust Bayesian viewpoint, pp 63–124
Berger JO (1985) Statistical decision theory and Bayesian analysis. Springer, New York
Berger JO (1990) Robust Bayesian analysis: sensitivity to the prior. J Stat Plan Inference 25:303–328
Berger JO, Sellke T (1987) Testing a point null hypothesis: the irreconcilability of p values and evidence. J Am Stat Assoc 82:112–122
Berger JO, Brown L, Wolpert R (1994) A unified conditional frequentist and Bayesian test for fixed and sequential simple hypothesis-testing. Ann Stat 22(4):1787–1807
Berger JO, Bernardo J, Sun D (2009) The formal definition of reference priors. Ann Stat 37(2):905–938
Bernardo JM (1979) Reference posterior distributions for Bayesian inference. J R Stat Soc B 41:113–147
Bickel DR (2011a) Estimating the null distribution to adjust observed confidence levels for genome-scale screening. Biometrics 67:363–370
Bickel DR (2011b) A predictive approach to measuring the strength of statistical evidence for single and multiple comparisons. Can J Stat 39:610–631
Bickel DR (2012a) Coherent frequentism: a decision theory based on confidence sets. Commun Stat Theory Methods 41:1478–1496
Bickel DR (2012b) Controlling the degree of caution in statistical inference with the Bayesian and frequentist approaches as opposite extremes. Electron J Statist 6:686–709
Bickel DR (2012c) Game-theoretic probability combination with applications to resolving conflicts between statistical methods. Int J Approx Reason 53:880–891
Bickel DR (2013) Simple estimators of false discovery rates given as few as one or two p-values without strong parametric assumptions. Stat Appl Genet Mol Biol 12:529–543
Bickel DR (2014a) A fiducial continuum from confidence sets to empirical Bayes set estimates as the number of comparisons increases. Working Paper, University of Ottawa, deposited in uO Research at http://hdl.handle.net/10393/31898
Bickel DR (2014b) Model fusion and multiple testing in the likelihood paradigm: shrinkage and evidence supporting a point null hypothesis. Working paper, University of Ottawa, deposited in uO Research at http://hdl.handle.net/10393/31897
Bickel DR (2014c) Small-scale inference: empirical Bayes and confidence methods for as few as a single comparison. Int Stat Rev 82:457–476
Bickel DR, Padilla M (2014) A prior-free framework of coherent inference and its derivation of simple shrinkage estimators. J Stat Plan Inference 145:204–221
Carlin BP, Louis TA (2009) Bayesian methods for data analysis, 3rd edn. Chapman and Hall/CRC, New York
Coletti C, Scozzafava R (2002) Probabilistic logic in a coherent setting. Kluwer, Amsterdam
Cover T, Thomas J (2006) Elements of information theory. Wiley, New York
Cox DR (2006) Principles of statistical inference. Cambridge University Press, Cambridge
Csiszár I (1991) Why least squares and maximum entropy? an axiomatic approach to inference for linear inverse problems. Ann Stat 19:2032–2066
DasGupta A, Studden W (1989) Frequentist behavior of robust Bayes estimates of normal means. Stat Decis 7:333–361
Dawid AP, Stone M (1982) The functional-model basis of fiducial inference (with discussion). Ann Stat 10:1054–1074
Efron B (2010a) Large-scale inference: empirical Bayes methods for estimation, testing, and prediction. Cambridge University Press, Cambridge
Efron B (2010b) Rejoinder to comments on B. Efron, “Correlated z-values and the accuracy of large-scale statistical estimates”. J Am Stat Assoc 105:1067–1069
Fisher RA (1973) Statistical methods and scientific inference. Hafner Press, New York
Fraser DAS (2004) Ancillaries and conditional inference. Stat Sci 19:333–351
Fraser DAS, Reid N (1990) Discussion: an ancillarity paradox which appears in multiple linear regression. Ann Stat 18:503–507
Gärdenfors P, Sahlin N-E (1982) Unreliable probabilities, risk taking, and decision making. Synthese 53:361–386
Genest C, Zidek JV (1986) Combining probability distributions: a critique and an annotated bibliography. Stat Sci 1:114–135
Gilboa I, Schmeidler D (1989) Maxmin expected utility with non-unique prior. J Math Econ 18(2):141–153
Good IJ (1952) Rational decisions. J R Stat Soc B 14:107–114
Good IJ (1983) Good thinking: the foundations of probability and its applications. G—reference, information and interdisciplinary subjects series. University of Minnesota Press, USA
Grünwald P, Dawid AP (2004) Game theory, maximum entropy, minimum discrepancy and robust Bayesian decision theory. Ann Stat 32:1367–1433
Hannig J (2009) On generalized fiducial inference. Stat Sinica 19:491–544
Harremoës P, Topsøe F (2001) Maximum entropy fundamentals. Entropy 3(3):191–226
Jaynes E (2003) Probability theory: the logic of science. Cambridge University Press, Cambridge
Jozani M, Marchand É, Parsian A (2012) Bayesian and robust bayesian analysis under a general class of balanced loss function. Stat Pap 53(1):51–60
Kakihara Y (1999) Abstract methods in information theory (series on multivariate analysis, volume 4). World Scientific, Singapore
Kass RE, Wasserman L (1996) The selection of prior distributions by formal rules. J Am Stat Assoc 91:1343–1370
Kracík J (2011) Combining marginal probability distributions via minimization of weighted sum of Kullback–Leibler divergences. Int J Approx Reason 52:659–671
Lavine M (1991) Sensitivity in Bayesian statistics: the prior and the likelihood. J Am Stat Assoc 86:396–399
Levi I (1986a) Hard choices: decision making under unresolved conflict. Cambridge University Press, Cambridge
Levi I (1986b) The paradoxes of Allais and Ellsberg. Econ Philos 2:23–53
Lindley DV (1957) A statistical paradox. Biometrika 44:187–192
Liu C, Aitkin M (2008) Bayes factors: prior sensitivity and model generalizability. J Math Psychol 52:362–375
Nadarajah S, Bityukov S, Krasnikov N (2015) Confidence distributions: a review. Stat Methodol 22:23–46
Neuhaus KL, von Essen R, Tebbe U, Vogt A, Roth M, Riess M, Niederer W, Forycki F, Wirtzfeld A, Maeurer W (1992) Improved thrombolysis in acute myocardial infarction with front-loaded administration of alteplase: results of the rt-PA-APSAC patency study (TAPS). J Am Coll Cardiol 19:885–891
Padilla M, Bickel D. R. (2012) Estimators of the local false discovery rate designed for small numbers of tests. Stat Appl Genet Mol Biol 11(5), art. 4
Paris JB (1994) The uncertain reasoner’s companion: a mathematical perspective. Cambridge University Press, New York
Paris J, Vencovská A (1997) In defense of the maximum entropy inference process. Int J Approx Reason 17(1):77–103
Pfaffelhuber E (1977) Minimax information gain and minimum discrimination principle. In: Csiszár I, Elias P (eds) Topics in information theory. Vol. 16 of Colloquia Mathematica Societatis János Bolyai. János Bolyai Mathematical Society and North-Holland, Amsterdam, pp 493–519
Polansky AM (2007) Observed confidence levels: theory and application. Chapman and Hall, New York
Robbins H (1951) Asymptotically subminimax solutions of compound statistical decision problems. Proc Second Berkeley Symp Math Stat Probab 1:131–148
Robinson GK (1991) That BLUP is a good thing: the estimation of random effects. Stat Sci 6:15–32
Savage LJ (1954) The foundations of statistics. Wiley, New York
Schweder T, Hjort NL (2002) Confidence and likelihood. Scand J Stat 29:309–332
Seidenfeld T (1988) Decision theory without independence or without ordering: what is the difference? Econ Philos 4(2):267–290
Seidenfeld T (2004) A contrast between two decision rules for use with (convex) sets of probabilities: Gamma-maximin versus E-admissibility. Synthese 140(1–2):69–88
Sellke T, Bayarri MJ, Berger JO (2001) Calibration of p values for testing precise null hypotheses. Am Stat 55:62–71
Sidak Z (1967) Rectangular confidence regions for means of multivariate normal distributions. J Am Stat Assoc 62(318):626–633
Singh K, Xie M, Strawderman WE (2007) Confidence distribution (CD)—distribution estimator of a parameter. IMS Lect Notes Monogr Ser 2007(54):132–150
Topsøe F (1979) Information theoretical optimization techniques. Kybernetika 15(1):8–27
Topsøe F (2004) Entropy and equilibrium via games of complexity. Phys A 340(1–3):11–31
Topsøe F (2007) Information theory at the service of science. In: Csiszár I, Katona GOH, Tardos G, Wiener G (eds) Entropy, search, complexity. Bolyai society mathematical studies. Springer, Berlin, pp 179–207
van Berkum E, Linssen H, Overdijk D (1996) Inference rules and inferential distributions. J Stat Plan Inference 49:305–317
Vidakovic B (2000) Gamma-minimax: a paradigm for conservative robust Bayesians. Robust Bayesian analysis. Springer, New York
Walley P (1991) Statistical reasoning with imprecise probabilities. Chapman and Hall, London
Weichselberger K (2000) The theory of interval-probability as a unifying concept for uncertainty. Int J Approx Reason 24(2–3):149–170
Weichselberger K (2001) Elementare Grundbegriffe einer allgemeineren Wahrscheinlichkeitsrechnung I: Intervallwahrscheinlichkeit als umfassendes Konzept. Physica-Verlag, Heidelberg
Westfall PH, Johnson WO, Utts JM (1997) A Bayesian perspective on the Bonferroni adjustment. Biometrika 84:419–427
Westfall PH (2010) Comment on B. Efron, “Correlated z-values and the accuracy of large-scale statistical estimates”. J Am Stat Assoc 105:1063–1066
Williamson J (2010) In defence of objective bayesianism. Oxford University Press, Oxford
Yuan B (2009) Bayesian frequentist hybrid inference. Ann Stat 37:2458–2501
Acknowledgments
The comments of the two anonymous reviewers and of the editor-in-chief are gratefully acknowledged for improving the clarity of presentation. In addition, I thank Xuemei Tang for providing the fruit-development microarray data. This research was partially supported by the Canada Foundation for Innovation, by the Ministry of Research and Innovation of Ontario, and by the Faculty of Medicine of the University of Ottawa.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bickel, D.R. Blending Bayesian and frequentist methods according to the precision of prior information with applications to hypothesis testing. Stat Methods Appl 24, 523–546 (2015). https://doi.org/10.1007/s10260-015-0299-6
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10260-015-0299-6
Keywords
- Confidence distribution
- Fiducial inference
- Imprecise probability
- Maximum entropy
- Multiple hypothesis testing
- Multiple comparison procedure
- Robust Bayesian analysis