The Interplay of Bayesian and Frequentist Analysis
The DOI matches its Crossref record; 106 of the 107 references it cites were found in a registry or an index, and 1 are not checked yet.
Abstract
Statistics has struggled for nearly a century over the issue of whether the Bayesian or frequentist paradigm is superior. This debate is far from over and, indeed, should continue, since there are fundamental philosophical and pedagogical issues at stake. At the methodological level, however, the debate has become considerably muted, with the recognition that each approach has a great deal to contribute to statistical practice and each is actually essential for full development of the other approach. In this article, we embark upon a rather idiosyncratic walk through some of these issues.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“However, roughly speaking, the Bayesian and frequentist approaches are usually considered the main paradigms [33,34].”
From Reverse engineering and identification in systems biology: strategies, perspectives and challenges (2014) · cites this paper as [33]
In the citing paper
“For models with tens to hundreds of parameters this is numerically demanding or even infeasible because high dimensional spaces cannot be sampled densely.”
From Likelihood based observability analysis and confidence intervals for predictions of dynamic models (2012) · cites this paper as [9]
In the citing paper
“Non-identifiable parameters cause the posterior PDF to be non-proper [28].”
From Joining forces of Bayesian and frequentist methodology: a study for inference in the presence of non-identifiability (2013) · cites this paper as [28]
Citations
Cites
Estimating the Dimension of a Model
Schwarz, G. (1978) · The Annals of Statistics · doi:10.1214/aos/1176344136 · Cited by 47,033 · cited 1×
“For instance, the popular Laplace approximation (cf. Schervish, 1995) and BIC (cf. Schwarz, 1978) are based on an asymptotic arguments.”
Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003) · doi:10.1201/9780429258480 · Cited by 27,276
Statistical Decision Theory and Bayesian Analysis
Berger, J. O. (1985) · Springer Series in Statistics · doi:10.1007/978-1-4757-4286-2 · Cited by 7,722 · cited 7×
“(See Berger, 1985a; Robert, 2001, for many examples and references.)”
InfluentialSelf-citationInterval Estimation for a Binomial Proportion
Brown, L. D., Cai, T. T., & DasGupta, A. (2001) · Statistical Science · doi:10.1214/ss/1009213286 · Cited by 3,538
Consistent Estimates Based on Partially Consistent Observations
Neyman, J., & Scott, E. L. (1948) · Econometrica · doi:10.2307/1914288 · Cited by 2,691 · cited 1×
“The classic example of this is the Neyman– Scott problem (Neyman and Scott, 1948), in which one observes”
Monte Carlo Statistical Methods
Robert, C. P., & Casella, G. (1999) · Springer Texts in Statistics · doi:10.1007/978-1-4757-3071-5 · Cited by 2,278
Bernardo, J. M., & Smith, A. F. M. (1994) · Wiley Series in Probability and Statistics · doi:10.1002/9780470316870 · Cited by 2,178
Rao, J. N. K. (2004) · Encyclopedia of Statistical Sciences · doi:10.1002/0471667196.ess0731 · Cited by 2,049
INADMISSIBILITY OF THE USUAL ESTIMATOR FOR THE MEAN OF A MULTIVARIATE NORMAL DISTRIBUTION
Stein, C. (1956) · Contribution to the Theory of Statistics · doi:10.1525/9780520313880-018 · Cited by 1,841
Bayesian Experimental Design: A Review
Chaloner, K., & Verdinelli, I. (1995) · Statistical Science · doi:10.1214/ss/1177009939 · Cited by 1,839 · cited 1×
“See Chaloner and Verdinelli (1995) and Dawid and Sebastiani (1999).”
Research works printed but not linked: 13
- Lehmann, E. L. and Casella, G. (1998). Theory of Point Estimation, 2nd ed. Springer, New York.
- Jeffreys, H. (1961). Theory of Probability, 3rd ed. Oxford Univ. Press.
- Robert, C. P. (2001). The Bayesian Choice, 2nd ed. Springer, New York.
- Wald, A. (1950). Statistical Decision Functions. Wiley, New York.
- Barnett, V. (1982). Comparative Statistical Inference, 2nd ed. Wiley, New York.
- Berger, J. (1985b). The frequentist viewpoint and conditioning. In Proceedings of the Berkeley Conference in Honor of Jerzy Neyman and Jack Kiefer (L. Le Cam and R. Olshen, eds.) 15--44. Wadsworth, Monterey, CA.
- Castellanos, M. E. (2002). Diagnóstico Bayesiano de modelos. Ph.D. dissertation, Univ. Miguel Hernández, Spain.
- de Finetti, B. (1970). Teoria delle Probabilità 1, 2. Einaudi, Torino. [English translations published (1974, 1975) as Theory of Probability 1, 2. Wiley, New York.]
- Diaconis, P. (1988b). Recent progress on de Finetti's notion of exchangeability. In Bayesian Statistics 3 (J. M. Bernardo, M. H. DeGroot, D. V. Lindley and A. F. M. Smith, eds.) 111--125. Oxford Univ. Press.
- Good, I. J. (1983). Good Thinking: The Foundations of Probability and Its Applications. Univ. Minnesota Press, Minneapolis.
- Hill, B. (1974). On coherence, inadmissibility and inference about many parameters in the theory of least squares. In Studies in Bayesian Econometrics and Statistics (S. Fienberg and A. Zellner, eds.) 555--584. North-Holland, Amsterdam.
- Lad, F. (1996). Operational Subjective Statistical Methods: A Mathematical, Philosophical and Historical Introduction. Wiley, New York.
- Paulo, R. (2002). Problems on the Bayesian/frequentist interface. Ph.D. dissertation, Duke Univ.
Cited by
A practical solution to the pervasive problems ofp values
Wagenmakers, E.-J. (2007) · Psychonomic Bulletin & Review · doi:10.3758/bf03194105 · Cited by 2,490
“The statistical world has always been dominated by two superpowers: the Bayesians and the frequentists (see, e.g., Bayarri & Berger, 2004; Berger, 2003; Christensen, 2005; R. T. Cox, 1946; Efron, 2005; Lindley & Phillips, 1976; Neyman, 1977).”
MethodPhilosophy and the practice of Bayesian statistics
Gelman, A., & Shalizi, C. R. (2013) · British Journal of Mathematical and Statistical Psychology · doi:10.1111/j.2044-8317.2011.02037.x · Cited by 825
“…the 10There is some controversy in the literature about whether posterior predictive checks have too little power to be useful statistical tools (Bayarri and Berger, 2000, 2004), how they might be modified to increase their power (Robins et al., 2000; Fraser and Rousseau, 2008), whether some…”
The case for objective Bayesian analysis
Berger, J. (2006) · Bayesian Analysis · doi:10.1214/06-ba115 · Cited by 728
“Bayarri and Berger (2004) review the vast literature showing that objective Bayesian methods are the most promising route to the unification of Bayesian and frequentist statistics.”
BackgroundSelf-citationColella, S., Yau, C., Taylor, J. M., Mirza, G., Butler, H., Clouston, P., Bassett, A. S., Seller, A., Holmes, C. C., & Ragoussis, J. (2007) · Nucleic Acids Research · doi:10.1093/nar/gkm076 · Cited by 616
“In instances where little or no substantive prior knowledge is available, the Objective Bayes approach provides a principled method to set parameters of the priors; such that the resulting Bayesian procedures possess good long-run frequency properties (29) (for general discussion of Objective Bayes see (21,22)).”
BackgroundMethodIntegrated Sensing and Communications: Recent Advances and Ten Open Challenges
Lu, S., Liu, F., Li, Y., Zhang, K., Huang, H., Zou, J., Li, X., Dong, Y., Dong, F., Zhu, J., et al. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3361173 · Cited by 516
“We remark that the CRB and BCRB/PCRB represent different perspectives in statistics, which constitute fundamental philosophical and cognitive issues [145].”
BackgroundMuff, S., Signer, J., & Fieberg, J. (2020) · Journal of Animal Ecology · doi:10.1111/1365-2656.13087 · Cited by 454
The best inflationary models after Planck
Martin, J., Ringeval, C., Trotta, R., & Vennin, V. (2014) · Journal of Cosmology and Astroparticle Physics · doi:10.1088/1475-7516/2014/03/039 · Cited by 301
“[19, 20] for a discussion of issues in Bayesianfrequentist calibrations).”
BackgroundAssessing Mediational Models: Testing and Interval Estimation for Indirect Effects
Biesanz, J. C., Falk, C. F., & Savalei, V. (2010) · Multivariate Behavioral Research · doi:10.1080/00273171.2010.498292 · Cited by 394
“For more extensive discussion of the differences between these perspectives see Bayarri and Berger (2004) .”
BackgroundStatistical Inference as Severe Testing: How to Get Beyond the Statistics Wars
Mayo, D. G. (2018) · doi:10.1017/9781107286184 · Cited by 376
Fair Statistical Communication in HCI
Dragicevic, P. (2016) · Human–Computer Interaction Series · doi:10.1007/978-3-319-26633-6_13 · Cited by 325
“This is true for exact confidence intervals (Bayarri and Berger, 2004, p.63) and bootstrap confidence intervals (Bååth, 2015).”
BackgroundGreenland, S. (2019) · The American Statistician · doi:10.1080/00031305.2018.1529625 · Cited by 313
“…if it is uniformly distributed when H and A are correct; in that case, for every α the rule “reject H when p ≤ α” will falsely reject H with frequency α (Bayarri and Berger 2004).2 Of course, issues of power must enter into choice of the test statistic from which p is computed, but those…”
BackgroundInfluentialSparse Modeling for Image and Vision Processing
Mairal, J. (2014) · now publishers, Inc. eBooks · doi:10.1561/9781680830095 · Cited by 322
Sparse Modeling for Image and Vision Processing
Ponce, J., Mairal, J., Bach, F., Image, S. M., Processing, V., Hal, H. I., & Supérieure, E. N. (2014) · Cited by 335
Smid, S. C., McNeish, D., Miočević, M., & van de Schoot, R. (2020) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2019.1577140 · Cited by 280
Bayesian Random-Effects Meta-Analysis Using the bayesmeta R Package
Röver, C. (2020) · Journal of Statistical Software · doi:10.18637/jss.v093.i06 · Cited by 310
“…between the results from frequentist and Bayesian analyses; often one may be considered a limiting or special case of the other, while interpretations remain somewhat different (Bartholomew 1965; Jaynes 1976; Lindley 1977; Severini 1991; Spiegelhalter et al. 1999; Bayarri and Berger 2004).”
BackgroundStatistical Foundations for Model-Based Adjustments
Greenland, S., & Pearce, N. (2015) · Annual Review of Public Health · doi:10.1146/annurev-publhealth-031914-122559 · Cited by 274
Negro, M. D., & Schorfheide, F. (2013) · Handbook of Economic Forecasting · doi:10.1016/b978-0-444-53683-9.00002-5 · Cited by 277
Machine Learning Methods for Property Prediction in Chemoinformatics:Quo Vadis?
Varnek, A., & Baskin, I. (2012) · Journal of Chemical Information and Modeling · doi:10.1021/ci200409x · Cited by 245
Ambiguity and the Bayesian Paradigm
Gilboa, I., & Marinacci, M. (2016) · Readings in Formal Epistemology · doi:10.1007/978-3-319-20451-2_21 · Cited by 249
Guidance on the use of the benchmark dose approach in risk assessment
EFSA Scientific Committee, More, S. J., Bampidis, V., Benford, D., Bragard, C., Halldorsson, T. I., Hernández‐Jerez, A. F., Bennekou, S. H., Koutsoumanis, K., Lambré, C., et al. (2022) · EFSA Journal · doi:10.2903/j.efsa.2022.7584 · Cited by 222
Are propensity scores really superior to standard multivariable analysis?
Biondi-Zoccai, G., Romagnoli, E., Agostoni, P., Capodanno, D., Castagno, D., D'Ascenzo, F., Sangiorgi, G., & Modena, M. G. (2011) · Contemporary Clinical Trials · doi:10.1016/j.cct.2011.05.006 · Cited by 227
“aspirin prevents recurrent myocardial infarction events, or cigarette smoking causes lung cancer) [2].”
BackgroundReverse engineering and identification in systems biology: strategies, perspectives and challenges
Villaverde, A. F., & Banga, J. R. (2014) · Journal of The Royal Society Interface · doi:10.1098/rsif.2013.0505 · Cited by 229
“However, roughly speaking, the Bayesian and frequentist approaches are usually considered the main paradigms [33,34].”
BackgroundWhy P Values Are Not a Useful Measure of Evidence in Statistical Significance Testing
Hubbard, R., & Lindsay, R. M. (2008) · Theory & Psychology · doi:10.1177/0959354307086923 · Cited by 212
“The upshot is that many researchers (e.g., Bayarri & Berger, 1999, 2000, 2004; Berger, 2003; Berger & Sellke, 1987; Gigerenzer, 1993; Goodman, 1993, 1999; Hubbard, 2004; Hubbard & Bayarri, 2003a, 2003b, 2005) state that the p value is routinely misinterpreted as a frequentist Type I error…”
BackgroundFarcomeni, A. (2008) · Statistical Methods in Medical Research · doi:10.1177/0962280206079046 · Cited by 225
Small sample mediation testing: Misplaced confidence in bootstrapped confidence intervals.
Koopman, J., Howe, M., Hollenbeck, J. R., & Sin, H.-P. (2015) · Journal of Applied Psychology · doi:10.1037/a0036635 · Cited by 201
Kim, N. J., Belland, B. R., & Walker, A. E. (2018) · Educational Psychology Review · doi:10.1007/s10648-017-9419-1 · Cited by 220
“Therefore, most Bayesians are reluctant to use frequentist hypothesis testing using p values because the observed data was not included in the posterior distribution (Babu 2012; Bayarri and Berger 2004; Kaweski and Nickeson 1997).”
MethodThe Posterior and the Prior in Bayesian Phylogenetics
Alfaro, M. E., & Holder, M. T. (2006) · Annual Review of Ecology, Evolution, and Systematics · doi:10.1146/annurev.ecolsys.37.091305.110021 · Cited by 215
“Objective Bayesian priors have a number of desirable properties: Many avoid problems that arise from priors that are sensitive to arbitrary choices (such as how to parameterize a model), and they often result in analyses that behave well from a frequentist viewpoint (Bayarri & Berger 2004).”
BackgroundConfidence, Likelihood, Probability: Statistical Inference with Confidence Distributions
Schweder, T., & Hjort, N. L. (2016) · Cambridge University Press eBooks · doi:10.1017/cbo9781139046671 · Cited by 179
Röver, C., Bender, R., Dias, S., Schmid, C. H., Schmidli, H., Sturtz, S., Weber, S., & Friede, T. (2021) · Research Synthesis Methods · doi:10.1002/jrsm.1475 · Cited by 179
Bayesian clinical trials in action
Jack Lee, J., & Chu, C. T. (2012) · Statistics in Medicine · doi:10.1002/sim.5404 · Cited by 200
“In-depth comparisons between the frequentist and Bayesian approaches can be found in the literature [5, 6].”
BackgroundBayesian sensitivity analysis for unmeasured confounding in observational studies
McCandless, L. C., Gustafson, P., & Levy, A. (2007) · Statistics in Medicine · doi:10.1002/sim.2711 · Cited by 181
“While average coverage probability is an unusual measure of frequentist performance, its relevance has been advocated elsewhere [19] because it characterizes performance of interval estimators in the setting in which they are actually used, namely to analyse data from different investigations, rather than data under h…”
BackgroundBias and Precision in Q ST Estimates: Problems and Some Solutions
O'Hara, R. B., & Merilä, J. (2005) · Genetics · doi:10.1534/genetics.105.044545 · Cited by 176
Vagnoli, M., Remenyte-Prescott, R., & Andrews, J. (2018) · Structural Health Monitoring · doi:10.1177/1475921717721137 · Cited by 171
Improved constraints on neutrino mixing from the T2K experiment with 3.13 × 10 21 protons on target
Abe, K., Akhlaq, N., Akutsu, R., Ali, A., Alt, C., Andreopoulos, C., Antonova, M., Aoki, S., Arihara, T., Asada, Y., et al. (2021) · Physical review. D/Physical review. D. · doi:10.1103/physrevd.103.112008 · Cited by 210
Information, Divergence and Risk for Binary ExperimentsPreprint
Reid, M. D., & Williamson, R. C. (2009) · arXiv (Cornell University) · doi:10.48550/arxiv.0901.0356 · Cited by 179
“There are several works (Barnett, 1999; Bayarri and Berger, 2004; Berger, 2003) that compare and contrast these.”
BackgroundIs the Jeffreys' scale a reliable tool for Bayesian model comparison in cosmology?
Nesseris, S., & García-Bellido, J. (2013) · Journal of Cosmology and Astroparticle Physics · doi:10.1088/1475-7516/2013/08/036 · Cited by 141
“The Jeffreys’ scale revisited In this section we will present results related to the Bayes factor Bij (see Refs [32], [33], [34], [35] and references there-in for more details) in the context of our simple model.”
MethodStatistical inference as severe testing how to get beyond the statistics wars
Mayo, D. G. (2018) · Cited by 159
Prior sensitivity analysis in default Bayesian structural equation modeling.
van Erp, S., Mulder, J., & Oberski, D. L. (2017) · Psychological Methods · doi:10.1037/met0000162 · Cited by 149
Universal darwinism as a process of bayesian inference
Campbell, J. O. (2016) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2016.00049 · Cited by 140
“Despite this interpretational dispute there has been some progress in uniting Bayesian and frequentist mathematics (Bayarri and Berger, 2004).”
The chaos in calibrating crop models
Wallach, D., Palosuo, T., Thorburn, P. J., Hochman, Z., Gourdain, E., Andrianasolo, F. N., Asseng, S., Basso, B., Buis, S., Crout, N. M. J., et al. (2020) · Environmental Modelling & Software · doi:10.1016/j.envsoft.2021.105206 · Cited by 135
Likelihood based observability analysis and confidence intervals for predictions of dynamic models
Kreutz, C., Raue, A., & Timmer, J. (2012) · BMC Systems Biology · doi:10.1186/1752-0509-6-120 · Cited by 134
“the parameter space to be sampled, is well-behaved [14].”
BackgroundCombining Expert Opinions in Prior Elicitation
Albert, I., Donnet, S., Guihenneuc‐Jouyaux, C., Low‐Choy, S., Mengersen, K. L., & Rousseau, J. (2012) · Bayesian Analysis · doi:10.1214/12-ba717 · Cited by 158
“This takes advantage of the frequentist property that in the long run we expect that, for an accurate expert, the elicited values ought to accurately reflect the true values across many elicitation topics (Bayarri and Berger 2004).”
BackgroundPost-Processing Posterior Predictive p Values
Hjort, N. L., Dahl, F. A., & Steinbakk, G. H. (2006) · Journal of the American Statistical Association · doi:10.1198/016214505000001393 · Cited by 149
“Important variations and alternative Bayesian p values were introduced by Bayarri and Berger (2000) and further discussed and analyzed by Robins, van der Vaart, and Ventura (2000), Bayarri and Berger (2004), and Bayarri and Castellanos (2004).”
Zhang, J., & Rowe, J. B. (2014) · Frontiers in Neuroscience · doi:10.3389/fnins.2014.00069 · Cited by 131
“We also used the group-level posteriors to compare two parameters in Bayesian methodology (Lindley, 1965; Berger and Bayarri, 2004; Kruschke, 2010, see section Data Processing and Analysis for details).”
MethodRaue, A., Kreutz, C., Theis, F. J., & Timmer, J. (2013) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2011.0544 · Cited by 119
“Nonidentifiable parameters cause the posterior PDF to be non-proper [28].”
BackgroundZhang, J., Rittman, T., Nombela, C., Fois, A., Coyle-Gilchrist, I., Barker, R. A., Hughes, L. E., & Rowe, J. B. (2016) · Brain · doi:10.1093/brain/awv331 · Cited by 108
“To date, 10 of the clinically diagnosed patients have died and been recruited to the Cambridge Brain Bank as part of a separate research programme: all 10 had neuropathological confirmation of the diagnosis.”
MethodEnTagRec: An Enhanced Tag Recommendation System for Software Information Sites
Wang, S., Lo, D., Vasilescu, B., & Serebrenik, A. (2014) · IEEE International Conference on Software Maintenance (ICSM) · doi:10.1109/icsme.2014.51 · Cited by 113
“ENTAGREC borrows from two opposite yet complementary lines of thought in the statistics community, Bayesian and frequentist [28], each with its own advantages [5], [13].”
BackgroundBayesian Analysis of Measurement Error Models Using Integrated Nested Laplace Approximations
Muff, S., Riebler, A., Held, L., Rue, H., & Saner, P. (2014) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1111/rssc.12069 · Cited by 121
“It is often of interest to study frequentist properties of Bayesian estimates (Bayarri and Berger, 2004).”
BackgroundRaue, A., Kreutz, C., Theis, F. J., & Timmer, J. (2016) · Cited by 101
Adaptive power priors with empirical Bayes for clinical trials
Gravestock, I., & Held, L. (2017) · Pharmaceutical Statistics · doi:10.1002/pst.1814 · Cited by 115
“Statistical Significance” and Statistical Reporting: Moving Beyond Binary
McShane, B. B., Bradlow, E. T., Lynch, J. G., & Meyer, R. J. (2024) · Journal of Marketing · doi:10.1177/00222429231216910 · Cited by 105
Measurement Uncertainty and Probability
Willink, R. (2013) · Cambridge University Press eBooks · doi:10.1017/cbo9781139135085 · Cited by 96
Objective priors for the bivariate normal model
Berger, J. O., & Sun, D. (2008) · The Annals of Statistics · doi:10.1214/07-aos501 · Cited by 90
Self-citationThe misuse and abuse of statistics in biomedical research
Thiese, M. S., Arnold, Z. C., & Walker, S. D. (2015) · Biochemia Medica · doi:10.11613/bm.2015.001 · Cited by 87
“There are many differences of opinion in methodological approaches, as exemplified by Frequentist and Bayesianist statistical methodology, or statistical estimation methods proposed by Fischer, Neyman, and Wald among others (11-15).”
MethodA critical evaluation of the current “p‐value controversy”
Wellek, S. (2017) · Biometrical Journal · doi:10.1002/bimj.201700001 · Cited by 92
“…for the majority of users of statistical methods is bought at the price that, except for fairly special cases of Bayesian analyses based on objective priors (cf. Bayarri and Berger, 2004), Bayesian inference does not allow one to control for risks of drawing erroneous conclusions from the data.”
BackgroundWeak Identifiability in Models for Mark-Recapture-Recovery Data
Giménez, O., Morgan, B. J. T., & Brooks, S. P. (2008) · Springer eBooks · doi:10.1007/978-0-387-78151-8_48 · Cited by 98
“... correlations between parameters in the posterior distribution, which in turn implies poor mixing in the MCMC samples and very slow convergence (Carlin and Louis 1996; Rannala 2002); even with large sample sizes, the likelihood may be unable to overcome the prior (Neath and Samaniego 1997); a too-informative prior…”
BackgroundObjective Bayesian Two Sample Hypothesis Testing for Online Controlled Experiments
Deng, A. (2015) · International World Wide Web Conference (WWW) · doi:10.1145/2740908.2742563 · Cited by 98
“As noted in Berger and Bayarri [3] and Berger and Wolpert [4] as the stopping rule principal, Bayesian methods with a genuine prior automatically support optional stopping, i.”
BackgroundNeutral noninformative and informative conjugate beta and gamma prior distributions
Kerman, J. (2011) · Electronic Journal of Statistics · doi:10.1214/11-ejs648 · Cited by 97
“Most practitioners are however probably more interested in the average performance rather than the worst possible performance over a sequence of trials [1, 3].”
BackgroundA Hierarchical Modelling Framework for Identifying Unusual Performance in Health Care Providers
Ohlssen, D. I., Sharples, L. D., & Spiegelhalter, D. J. (2007) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.2007.00487.x · Cited by 82
“When using full cross-validation this type of ‘partial’ p-value is, asymptotically, a proper p-value with a uniform distribution on (0,1) (Robins et al., 2000; Bayarri and Berger, 2004).”
BackgroundMethodInfluentialContinuous Monitoring of A/B Tests without Pain: Optional Stopping in Bayesian Testing
Deng, A., Lu, J., & Chen, S. (2016) · IEEE International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa.2016.33 · Cited by 96
“This issue is also discussed in Berger and Berry [3] and later Berger and Bayarri [2] referred to the idea as the “stopping rule principle” and said “once the data have been obtained, the reasons for stopping experimentation should have no bearing on the evidence reported about unknown model parameters.”
BackgroundIntelligent management of intelligence agencies: Beyond accountability ping-pong.
Tetlock, P. E., & Mellers, B. A. (2011) · American Psychologist · doi:10.1037/a0024285 · Cited by 71
“The third objection ultimately rests, however, on a narrow frequentist conception of probability that discounts the value of a subjectivist or Bayesian conception of uncertainty (Bayarri & Berger, 2004) and the benefits of imperfect quantification.”
BackgroundEvaluating DSGE model forecasts of comovements
Herbst, E. P., & Schorfheide, F. (2012) · Journal of Econometrics · doi:10.1016/j.jeconom.2012.06.008 · Cited by 71
Sparse Modeling for Image and Vision Processing
Mairal, J., Bach, F., & Ponce, J. (2014) · Foundations and Trends® in Computer Graphics and Vision · doi:10.1561/0600000058 · Cited by 267
“…from the point of view of a Bayesian statistician, who is usually not interested in point estimates, but in the full posterior distribution [see Bayarri and Berger, 2004].5 In fact, the formulation (2.9) discards any uncertainty regarding the model parameters D and A and would be called a…”
BackgroundInfluentialIncreased scientific rigor will improve reliability of research and effectiveness of management
Sells, S. N., Bassing, S. B., Barker, K. J., Forshee, S. C., Keever, A. C., Goerz, J. W., & Mitchell, M. S. (2018) · Journal of Wildlife Management · doi:10.1002/jwmg.21413 · Cited by 71
Bayesian Probability and Statistics in Management Research: A New Horizon
Zyphur, M. J., & Oswald, F. L. (2012) · Journal of Management · doi:10.1177/0149206312463183 · Cited by 70
“…for estimation and inference (Efron, 2005, 2010; Little, 2006), and there are many situations where both techniques are useful and can inform one another (see Bayarri & Berger, 2004; Berger, 2003; Berger, Boukai, & Wang, 1997; Berger, Brown, & Wolpert, 1994; Selke, Bayarri, & Berger, 2001).”
BackgroundSanz‐Cuesta, T., Escortell‐Mayor, E., del Cura-González, I., Martín‐Fernández, J., Fuertes, R. R., Garrido-Elustondo, S., Mariño-Suárez, J. E., Álvarez-Villalba, M., Gómez‐Gascón, T., González-García, I., et al. (2020) · BMJ Open · doi:10.1136/bmjopen-2019-033687 · Cited by 70
Schmitz, J. M., Rathnayaka, N., Green, C. E., Moeller, F. G., Dougherty, A. E., & Grabowski, J. (2012) · Frontiers in Psychiatry · doi:10.3389/fpsyt.2012.00077 · Cited by 65
“Together, these two complementary approaches can provide a more accurate parameter estimate when the sample size is small (for a full description, see Bayarri and Berger, 2004; Green et al., 2009).”
BackgroundMethodDiagnostics of prior-data agreement in applied Bayesian analysis
Bousquet, N. (2008) · Journal of Applied Statistics · doi:10.1080/02664760802192981 · Cited by 62
“As Bayarri and Berger 8 9 recommend, p -values must be carefully used; the understanding of the result could thus be mistaken for industrial analysts that are not statisticians.”
BackgroundBayesian spatio-temporal approach to identifying fish nurseries by validating persistence areas
Paradinas, I., Conesa, D., Pennino, M. G., Muñoz, F., Fernández, A. M., López‐Quílez, A., & Bellido-Millán, J. M. (2015) · Marine Ecology Progress Series · doi:10.3354/meps11281 · Cited by 66
“In order to ex - press our lack of a priori information about the para - meters, we adopted an objective Bayesian approach, an attempt to unify frequentist and Bayesian statistics (Bayarri & Berger 2004).”
MethodEvaluating Model Performance in Evolutionary Biology
Brown, J. M., & Thomson, R. C. (2018) · Annual Review of Ecology Evolution and Systematics · doi:10.1146/annurev-ecolsys-110617-062249 · Cited by 62
Computational Advances for and from Bayesian Analysis
Andrieu, C., Arnaud Doucet, B., & Robert, C. P. (2004) · Statistical Science · doi:10.1214/088342304000000071 · Cited by 61
“While the overall usefulness of Bayes factors in Bayesian testing may be argued (Kass and Raftery, 1995; Bayarri and Berger, 2004; Walker, 2004), they are nonetheless part of the standard Bayesian toolbox, if only as a reference value, for the comparison of modelsM1 andM2.”
BackgroundGreenland, S. (2022) · Scandinavian Journal of Statistics · doi:10.1111/sjos.12625 · Cited by 62
“The description of P-values as compatibility measures can be seen in Box (1980, p. 387), Bayarri & Berger (2000, 2004), Robins et al. (2000) and Greenland (2019a), and is anticipated by this passage in Fisher (1935, p. 207; emphasis added): “If we consider a series of hypotheses with different…”
BackgroundInfluentialParadinas, I., Conesa, D., López‐Quílez, A., & Bellido-Millán, J. M. (2017) · Spatial Statistics · doi:10.1016/j.spasta.2017.08.001 · Cited by 61
Reconnecting p -Value and Posterior Probability Under One- and Two-Sided Tests
Shi, H., & Yin, G. (2020) · The American Statistician · doi:10.1080/00031305.2020.1717621 · Cited by 65
Hand, D. J. (2015) · Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences · doi:10.1098/rsta.2014.0252 · Cited by 55
Posterior propriety and admissibility of hyperpriors in normal hierarchical models
Berger, J. O., Strawderman, W. E., & Tang, D. (2005) · The Annals of Statistics · doi:10.1214/009053605000000075 · Cited by 55
“The most successful approach to evaluation of objective improper priors has been to study the frequentist properties of the ensuing Bayes procedures (see [3] for discussion and many references).”
BackgroundSelf-citationDeng, Y., Barros, A., & Grall, A. (2015) · IEEE Transactions on Reliability · doi:10.1109/tr.2015.2462353 · Cited by 58
“The choice between parametric and non-parametric consideration has been debated for over a century [7].”
BackgroundInterval estimation in the presence of nuisance parameters. 1. Bayesian approachPreprint
Heinrich, J. G., Blocker, C., Conway, J., Demortier, L., Lyons, L., Giovanni Punzi, F., & Sinervo, P. (2004) · arXiv (Cornell University) · doi:10.48550/arxiv.physics/0409129 · Cited by 54
“(4) The limit is now obtained by integrating this posterior p.d.f. for s until we achieve the required fraction β of the total integral from zero to infinity.”
BackgroundThere Is Individualized Treatment. Why Not Individualized Inference?
Liu, K., & Meng, X. (2016) · Annual Review of Statistics and Its Application · doi:10.1146/annurev-statistics-010814-020310 · Cited by 59
Comparison of REML and Gibbs sampling estimates of multi-trait genetic parameters in Scots pine
Waldmann, P., & Ericsson, T. (2006) · Theoretical and Applied Genetics · doi:10.1007/s00122-006-0246-x · Cited by 53
“The long-lasting and sometimes infected controversy between practicians of standard frequentist and Bayesian statistics is likely to continue for some time, but methodologically both paradigms seem to be converging (Blasco 2001; Bayarri and Berger 2004).”
BackgroundMethodBrittleness of Bayesian inference under finite information in a continuous world
Owhadi, H., Scovel, C., & Sullivan, T. (2015) · Electronic Journal of Statistics · doi:10.1214/15-ejs989 · Cited by 66
“This approach, oftentimes described as “frequentist” [55, 11], could be used to validate Bayesian calculations [43].”
MethodAn Overview of the Adaptive Designs Accelerating Promising Trials Into Treatments (ADAPT-IT) Project
Meurer, W. J., Lewis, R. J., Tagle, D. A., Fetters, M. D., Legocki, L. J., Berry, S. M., Connor, J., Durkalski, V., Elm, J., Zhao, W., et al. (2012) · Annals of Emergency Medicine · doi:10.1016/j.annemergmed.2012.01.020 · Cited by 51
Boughton, D. A., GIBSON, M., YEDOR, R., & Kelley, E. (2007) · Freshwater Biology · doi:10.1111/j.1365-2427.2007.01772.x · Cited by 56
“For insight on contrasts between Bayesian and frequentist statistics, see Wolpert (2004) and Bayarri & Berger (2004).”
BackgroundMayo, D. G., & Hand, D. J. (2022) · Synthese · doi:10.1007/s11229-022-03692-0 · Cited by 57
“…reason that even Bayesians turn to simple significance tests, if they want to check their assumptions, is that such tests do not require specifying an alternative, but only a single, appropriately chosen, null hypothesis (e.g., Bayarri & Berger, 2004; Box, 1983; Gelman, 2011; Gelman&Shalizi, 2013).”
BackgroundMethodHeinrich, J., & Lyons, L. (2007) · Annual Review of Nuclear and Particle Science · doi:10.1146/annurev.nucl.57.090506.123052 · Cited by 48
Mendelian randomization incorporating uncertainty about pleiotropy
Thompson, J. R., Minelli, C., Bowden, J., Del Greco, F. M., Gill, D., Jones, E. M., Shapland, C. Y., & Sheehan, N. A. (2017) · Statistics in Medicine · doi:10.1002/sim.7442 · Cited by 45
“Although the relevance of frequentist properties to Bayesian analysis is debatable [26], the gain in model averaging comes from a bias-variance trade-off that produces better prediction with smaller root mean square error [27, 18].”
BackgroundConstruction schedule risk analysis – a hybrid machine learning approach
Fitzsimmons, J., Lu, R., Hong, Y., & Brilakis, I. (2022) · Journal of Information Technology in Construction · doi:10.36680/j.itcon.2022.004 · Cited by 46
“These factors can be included in an Empirical Bayesian Network, leading themselves to a data-driven approach.”
BackgroundInfluentialThe three stages of religious decline around the world
Stolz, J., de Graaf, N. D., Hackett, C., & Antonietti, J. (2025) · Nature Communications · doi:10.1038/s41467-025-62452-z · Cited by 55
“We follow Gelmanand Shalizi in their view that both frequentist and Bayesian methods can be combined pragmatically in a hypothetico-deductive scienti fi c framework 42,43 .”
Inference and Decision Making for 21st-Century Drug Development and Approval
Ruberg, S. J., Harrell, F. E., Gamalo, M., LaVange, L. M., Lee, J. J., Price, K., & Peck, C. C. (2019) · The American Statistician · doi:10.1080/00031305.2019.1566091 · Cited by 48
“There is abundant literature comparing and contrasting both approaches (Bayarri and Berger 2004; Christensen 2005; Inoue et al. 2005).”
Cognitive aging and verbal labeling in continuous visual memory
Forsberg, A., Johnson, W., & Logie, R. H. (2020) · Memory & Cognition · doi:10.3758/s13421-020-01043-3 · Cited by 47
Evaluating methods of calculating measurement uncertainty
Hall, B. D. (2008) · Metrologia · doi:10.1088/0026-1394/45/2/n01 · Cited by 45
Credible Confidence: A Pragmatic View on the Frequentist vs Bayesian Debate
Albers, C. J., Kiers, H. A. L., & van Ravenzwaaij, D. (2018) · Collabra Psychology · doi:10.1525/collabra.149 · Cited by 48
“For instance, the paper by Bayarri and Berger (2004) starts with “Statisticians should readily use both Bayesian and frequentist ideas.”
BackgroundBayesian Astrostatistics: A Backward Look to the Future
Loredo, T. J. (2012) · Springer series in astrostatistics · doi:10.1007/978-1-4614-3508-2_2 · Cited by 42
“ubset of parameters, ψ, and uninteresting “nuisance” parameters, η. The uncertainty in ψ(with the ηuncertainty fully propagated) is quantified by the marginal density, p(ψ|Dobs)= Z dηp(ψ,η|Dobs) || M. (2) • Predicting future data, D′, with the posterior predictive distribution, p(D′|Dobs)= Z dθp(D′|θ)p(θ|Dobs) ||M, (3)…”
BackgroundMethodInfluentialPosterior propriety in Bayesian extreme value analyses using reference priors
Northrop, P. J., & Attalides, N. (2015) · Statistica Sinica · doi:10.5705/ss.2014.034 · Cited by 42
“vels, placing independent normal priors on , log˙and ˘with respective variances 104;104 and 100. However, one needs to check that the posterior is not sensitive to the choice of proper prior and, as Bayarri and Berger (2004) note \...these posteriors will essentially be meaningless if the limiting improper objective p…”
Backgroundpytc : Open-Source Python Software for Global Analyses of Isothermal Titration Calorimetry Data
Duvvuri, H., Wheeler, L. C., & Harms, M. J. (2018) · Biochemistry · doi:10.1021/acs.biochem.7b01264 · Cited by 42
Power priors based on multiple historical studies for binary outcomes
Gravestock, I., & Held, L. (2018) · Biometrical Journal · doi:10.1002/bimj.201700246 · Cited by 45
Bayesian nonparametric model selection and model testing
Karabatsos, G. (2005) · Journal of Mathematical Psychology · doi:10.1016/j.jmp.2005.07.003 · Cited by 44
In pursuit of a fit-for-purpose uncertainty guide
White, D. R. (2016) · Metrologia · doi:10.1088/0026-1394/53/4/s107 · Cited by 42
Saz‐Lara, A., Cavero‐Redondo, I., Álvarez‐Bueno, C., Notario‐Pacheco, B., Reina‐Gutiérrez, S., Sequí‐Domínguez, I., Ruiz, J. R., & Martínez‐Vizcaíno, V. (2021) · European Journal of Cardiovascular Nursing · doi:10.1093/eurjcn/zvab022 · Cited by 39
What’s more general than a whole population?
Alexander, N. D. E. (2015) · Emerging Themes in Epidemiology · doi:10.1186/s12982-015-0029-4 · Cited by 43
“Moreover, many Bayesian problems allow ‘matching’ priors which replicate frequentist results [16].”
Background
Metadata
- Authors
- Bayarri, M. J., & Berger, J. O.
- ORCID
- Berger, J. O.
- Venue
- Statistical Science
- Publisher
- Institute of Mathematical Statistics
- Published
- 2004-02-01
- Volume
- 19
- Issue
- 1
- Pages
- 58-80
- ISSN
- 0883-4237 (ISSN-L)
- 2168-8745
- OpenAlex
- W647244699
- Cited by
- 525
- References
- 94
- Citations by year
- 2026: 11 · 2025: 13 · 2024: 24 · 2023: 23 · 2022: 24 · 2021: 34 · 2020: 34 · 2019: 20 · 2018: 49 · 2017: 26 · 2016: 34 · 2015: 28 · 2014: 40 · 2013: 21 · 2012: 33
- FWCI
- 14.90
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Statistical Science (publishedVersion)
- Project Euclid (Cornell University) (submittedVersion)
- Quality Engineering
- Funders
- National Science Foundation
- Ministerio de Ciencia y Tecnología
- Division of Mathematical Sciences
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Probability and Statistical Research
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Keywords
- Bayesian inference, frequentist inference, philosophy of statistics, statistical methodology
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.