Prior distributions for variance parameters in hierarchical models (comment on article by Browne and Draper)
Open accessThe DOI matches its Crossref record; 42 of the 42 references it cites were found in a registry or an index; 1 of the 23 sources its claims cite were checked for support.
Abstract
Various noninformative prior distributions have been suggested for scale parameters in hierarchical models. We construct a new folded-noncentral-$t$ family of conditionally conjugate priors for hierarchical standard deviation parameters, and then consider noninformative and weakly informative priors in this family. We use an example to illustrate serious problems with the inverse-gamma family of "noninformative" prior distributions. We suggest instead to use a uniform prior on the hierarchical standard deviation, using the half-$t$ family when the number of groups is small and in other settings where a weakly informative prior is desired. We also illustrate the use of the half-$t$ family for hierarchical modeling of multiple variance parameters such as arise in the analysis of variance.
Overview
CoChat Summary
TL;DR
Gelman argues that inverse-gamma “noninformative” priors for hierarchical variance parameters can badly distort posterior inference, especially when the group-level variance is near zero or the number of groups is small. He recommends a uniform prior on the hierarchical standard deviation when groups are not few, and half-t/half-Cauchy weakly informative priors otherwise, developed within a folded-noncentral-t conditionally conjugate family.
Key findings
- The inverse-gamma(ε, ε) family is not reliably noninformative: as ε → 0 it yields an improper posterior, and in the 8-schools example inverse-gamma(1, 1) and inverse-gamma(0.001, 0.001) priors on σ²_α constrained posterior inferences and increased shrinkage relative to a uniform prior on σ_α.
- A uniform prior on σ_α performed well for the 8-schools data (J = 8), but with only J = 3 schools it produced an extremely long right tail and undershrinkage; a half-Cauchy prior with scale A = 25 removed much of that unrealistic upper tail.
- For multiple variance components, Gelman recommends a hierarchical half-Cauchy model with scale estimated from data; in the 5 × 5 × 2 split-plot latin square Anova with nine variance components, the Cauchy scale parameter A was estimated at 1.8 with a 95% posterior interval of [0.5, 5.1].
- The folded-noncentral-t family, obtained via a redundant multiplicative parameterization, provides conditionally conjugate priors for σ_α and includes the improper uniform density (ν = −1) and proper half-Cauchy (ν = 1) as special cases.
Methods
- Conceptual/comment paper using a simple two-level normal hierarchical model: y_ij ∼ N(μ + α_j, σ²_y), α_j ∼ N(0, σ²_α), focusing on the prior for σ_α.
- Illustrates with the 8-schools example (J = 8, 6000 Bugs simulation draws), a 3-schools subset, and a 5 × 5 × 2 split-plot latin square Anova with nine variance components.
- Uses a parameter-expanded form y_ij ∼ N(μ + ξη_j, σ²_y), η_j ∼ N(0, σ²_η), with σ_α = |ξ|σ_η, to construct and compare uniform, inverse-gamma, half-Cauchy, and hierarchical half-Cauchy priors.
Limitations
- The paper states that noninformative and weakly informative priors are inherently provisional: after fitting, one should inspect the posterior and revise the prior if it does not make sense.
- The paper notes that the uniform prior on σ_α has a positive miscalibration and yields an improper posterior for J = 1 or 2; inverse-gamma inferences are sensitive to ε; and half-Cauchy scale choices are weakly informative rather than fully objective.
- Evident from the text: the recommendations rest on illustrative examples and conceptual argument rather than a broad simulation study or systematic empirical comparison, and extensions to nonlinear, multivariate, and more complex hierarchical models are left for further work.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 14 Not checked · 56 Own results and interpretations
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“Similarly, there are no always-nonnegative classical unbiased estimators of σ α or σ 2 α in the hierarchical model.”
- Also stated in: § 7.1 Prior distributions for variance parameters · plain
“Similarly, there are no alwaysnonnegative classical unbiased estimators of σ α or σ 2 α in the hierarchical model.”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 2.4 Calibration
What the sources say
- Could not verifycites Bickel and Blackwell (1967) matched to “A Note on Bayes Estimates” (1967) sourceExperimental — accuracy not yet measured
The introduction (p. 1) and Theorem 1.3 (p. 3) concern when an unbiased Bayes estimate must have zero Bayes risk. Searches for hierarchical models, σₐ, σ²ₐ, and always-nonnegative classical unbiased estimators found no passage addressing the claimed nonexistence result. (LLM, conf=0.77)
- Not checked — no full text of the sourcecites Meng and Zaslavsky (2002) matched to “Single observation unbiased priors” (2002) source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 7.1 Prior distributions for variance parameters · plain
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“If J is low, the uniform prior density tends to lead to high estimates of σ α , as discussed in Section 5.2.”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 7.1 Prior distributions for variance parameters
hedgedstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In contrast, the hierarchical half-Cauchy model performs a great deal of shrinkage, especially of the high ranges of the intervals.”
- Also stated in: § Bugs model: a half-Cauchy prior distribution on sigma.theta is induced # using a normal prior on xi and an inverse-gamma on tau.eta model { for (j in 1:J){ # J = the number of schools y[j] ~dnorm (theta[j], tau.y[j]) # data model: the likelihood theta[j] <-mu.theta+ xi*eta[j] tau.y[j] <-pow(sigma.y[j],-2) } xi ~dnorm (0, tau.xi) tau.xi <-pow(prior.scale,-2) for (j in 1:J){ eta[j] ~dnorm (0, tau.eta) # hierarchical model for theta } tau.eta ~dgamma (.5, .5)# chi^2 with 1 d.f.sigma.theta<-abs(xi)/sqrt(tau.eta)# cauchy = normal/sqrt(chi^2) mu.theta ~dnorm (0.0, 1.0E-6) # noninformative prior on mu } · plain
“The half-Cauchy model gives much sharper inferences, using the partial pooling that comes with fitting a hierarchical model.”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 6.2 Superpopulation and finite-population standard deviations
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § Bugs model: a half-Cauchy prior distribution on sigma.theta is induced # using a normal prior on xi and an inverse-gamma on tau.eta model { for (j in 1:J){ # J = the number of schools y[j] ~dnorm (theta[j], tau.y[j]) # data model: the likelihood theta[j] <-mu.theta+ xi*eta[j] tau.y[j] <-pow(sigma.y[j],-2) } xi ~dnorm (0, tau.xi) tau.xi <-pow(prior.scale,-2) for (j in 1:J){ eta[j] ~dnorm (0, tau.eta) # hierarchical model for theta } tau.eta ~dgamma (.5, .5)# chi^2 with 1 d.f.sigma.theta<-abs(xi)/sqrt(tau.eta)# cauchy = normal/sqrt(chi^2) mu.theta ~dnorm (0.0, 1.0E-6) # noninformative prior on mu } · plain
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This prior distribution is even more sharply peaked near zero and further distorts posterior inferences, with the problem arising because the marginal likelihood for σ α remains high near zero.”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 5.1 Noninformative prior distributions for the 8-schools problem
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The uniform prior distribution seems fine for the 8-school analysis”
- Also stated in: § 5.2 Weakly informative prior distribution for the 3-schools problem · plain
“the default uniform prior gave reasonable results”
- Also stated in: § 7.1 Prior distributions for variance parameters · hedged
“Figure 1 illustrates the generally robust properties of the uniform prior density on σ α .”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 5.2 Weakly informative prior distribution for the 3-schools problem
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Also stated in: § 5.2 Weakly informative prior distribution for the 3-schools problem · plain
26 more claims
Create a free account to see all 71 claims and every source verdict.
- It may state two separate claims as one. 34
- Probably not a substantial claim. 12
- The quote only partly states this claim. 1
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Because the uniform distribution tends to unduly favor presence of heterogeneity in discrimination and calibration estimates across studies,58,59”
From A framework for meta-analysis of prediction model studies with binary and time-to-event outcomes (2019) · cites this paper as [59]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 7.1 Prior distributions for variance parameters
“we see conditional conjugacy in the wider class of half-t distributions on σ α , which include the uniform and half-Cauchy densities on σ α (as well as inverse-gamma on σ 2 α ) as special cases.”
§ Prior distributions for variance parameters in hierarchical models(Comment on Article by Browne and Draper) > 3.2 Folded-noncentral-t prior distribution for σ α
“In the limit that the denominator is specified exactly, we have a folded normal distribution; conversely, specifying the numerator exactly yields the square-root-inverse-χ 2 distribution for σ α , as in Section 3.1.”
Citations
Cites
Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003) · doi:10.1201/9780429258480 · Cited by 27,276
Self-citationKass, R. E., & Raftery, A. E. (1995) · Journal of the American Statistical Association · doi:10.1080/01621459.1995.10476572 · Cited by 12,558
Sampling-based approaches to calculating marginal densities
Gelfand, A. E., & Smith, A. F. M. (1990) · Journal of the American Statistical Association · doi:10.1080/01621459.1990.10476213 · Cited by 6,606 · cited 1×
“Typically, enough data will be available to estimate µ and σy that one can use any reasonable noninformative prior distribution—for example, p(µ, σy) ∝ 1 or p(µ, log σy) ∝ 1.”
Introducing Multilevel Modeling
Kreft, I., & de Leeuw, J. (1998) · doi:10.4135/9781849209366 · Cited by 2,997
POSTERIOR PREDICTIVE ASSESSMENT OF MODEL FITNESS VIA REALIZED DISCREPANCIES
Gelman, A., Meng, X., & Stern, H. S. (1996) · Cited by 2,638
Self-citationBayes and Empirical Bayes Methods for Data Analysis
(2001) · Technometrics · doi:10.1198/tech.2001.s608 · Cited by 2,098
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 Approaches to Clinical Trials and Health‐Care Evaluation
Spiegelhalter, D. J., Abrams, K. R., & Myles, J. P. (2003) · Wiley eBooks · doi:10.1002/0470092602 · Cited by 1,673
Parametric empirical Bayes inference: theory and applications
Morris, C. N. (1983) · Journal of the American Statistical Association · doi:10.1080/01621459.1983.10477920 · Cited by 1,448 · cited 1×
“On the theoretical side, hierarchical models allow a more “objective” approach to inference by estimating the parameters of prior distributions from data rather than requiring them to be specified using subjective information (see James and Stein, 1960, Efron and Morris, 1975, and Morris, 1983).”
The Selection of Prior Distributions by Formal Rules
Kass, R. E., & Wasserman, L. (1996) · Journal of the American Statistical Association · doi:10.1080/01621459.1996.10477003 · Cited by 1,308
Research works printed but not linked: 5
- Box, G. E. P. and Tiao, G. C. (1973). Bayesian Inference in Statistical Analysis. Reading, Mass.: Addison-Wesley.
- James, W. and Stein, C. (1960). "Estimation with quadratic loss." In Neyman, J. (ed.), Proceedings of the Fourth Berkeley Symposium, volume 1, 361–380. Berkeley: University of California Press.
- Johnson, N. L. and Kotz, S. (1972). Distributions in Statistics. New York: Wiley. 4 vols.
- Liu, C. (2001). "Bayesian analysis of multivariate probit models. Discussion of “The art of data augmentation” by D. A. van Dyk and X. L. Meng." Journal of Computational and Graphical Statistics, 10: 75–81.
- Raudenbush, S. W. and Bryk, A. S. (2002). Hierarchical Linear Models. Thousand Oaks, Calif.: Sage., second edition.
Cited by
Items 2,801–2,900 of 4,089
Smith, B. J., & Beichel, R. (2017) · Statistical Methods in Medical Research · doi:10.1177/0962280217741334 · Cited by 3
“Uniforms were employed for the standard deviation parameters to represent weakly informative priors, as suggested by Gelman.(11) Semiconjugate inverse-gammas for the variance forms of these parameters are additionally available in our software implementation of the model to accommodate more informative priors.”
MethodLiu, S., Ryu, D., Western, A., et al. (2017) · MODSIM · doi:10.36334/modsim.2017.l22.liu · Cited by 3
“Noninformative uniform priors were assigned to standard deviations (σ ,σβ) and non-informative normal distribution were assigned as the prior distribution for regression coefficients (α, γ0, γn) that describe the data model and the process model (Gelman, 2006):”
MethodInfluentialBayesian Point Process Modeling to Quantify Geographic Regions of Excess Stillbirth Risk
Zahrieh, D., Oleson, J., & Romitti, P. A. (2018) · Geographical Analysis · doi:10.1111/gean.12174 · Cited by 3
Bayesian approach, traditional method, and mixed models for multienvironment trials of soybean
da Silva, A. J., Sanches, A., Andrade, A. C. B., et al. (2018) · Pesquisa Agropecuária Brasileira · doi:10.1590/s0100-204x2018001000002 · Cited by 3
Liu, Y., & Chen, Q. (2019) · Journal of Survey Statistics and Methodology · doi:10.1093/jssam/smz016 · Cited by 3
“Half-Cauchy priors suggested in Gelman (2006) are also considered with sb Cauchy ð0; 1Þþ in the simulation studies and implemented using probabilistic programming language Stan (Carpenter, Gelman, Hoffman, Lee, Goodrich, et al. 2017).”
MethodA Quantitative Framework to Inform Extrapolation Decisions in Children
Wadsworth, I., Hampson, L. V., Jaki, T., Sills, G. J., Marson, A. G., & Appleton, R. (2019) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/rssa.12532 · Cited by 3
“Gelman (2006) showed that inverse-gamma(", ") priors with " ≈ 0 are informative for variance parameters in hierarchical models and suggested that inverse Wishart prior distributions for covariance matrices incur similar issues.”
Iyengar, S., Lee, S., Irwin, D., et al. (2021) · ACM/IMS Transactions on Data Science · doi:10.1145/3406961 · Cited by 3
“positive as they represent increase in energy per unit temperature. The parameters of the gaussian priors are scaled to our setting and selected based on the recommendations provided by Gelman et al. [24]. To simplify our building model, we assume that the parameters are independent, i.e., the heating, cooling and the…”
Mui, M. L., Famula, T. R., Henthorn, P. S., et al. (2020) · PLoS ONE · doi:10.1371/journal.pone.0239542 · Cited by 3
“5), as recommended for weakly informative prior distributions [16].”
BackgroundSpatio-temporal parse network-based trajectory modeling on the dynamics of criminal justice system
Yu, H., Jiang, S., & Huang, H. (2021) · Journal of Applied Statistics · doi:10.1080/02664763.2021.1887101 · Cited by 3
The utility of Bayesian data reconciliation for separations
Koermer, S., & Noble, A. (2021) · Minerals Engineering · doi:10.1016/j.mineng.2021.106837 · Cited by 2
Tudor, J., & Wall, M. (2021) · Journal of Elections Public Opinion and Parties · doi:10.1080/17457289.2021.1972305 · Cited by 3
“Our preferred model uses half-Cauchy priors for the variance parameters, as recommended for MCMC multi-level modelling (Gelman 2006).”
MethodA Time Series Decomposition Algorithm Based on Gaussian Processes
Bilancia, M., Manca, F., & Sansaro, G. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86973-1_41 · Cited by 3
Lee, S. Y., Munafo, A., Girard, P., & Goteti, K. (2021) · Contemporary Clinical Trials · doi:10.1016/j.cct.2021.106657 · Cited by 3
Xu, S. (2012) · Springer eBooks · doi:10.1007/978-0-387-70807-2_2 · Cited by 3
Aghabazaz, Z., Kazemi, I., & Nematollahi, A. (2022) · Statistical Modelling · doi:10.1177/1471082x221080488 · Cited by 2
“We adopt A time-varying GARCH mixed-effects model 9 as proposed by Gelman (2006) and Polson and Scott (2012).”
BackgroundMethodGodoy, L. D. C., Prates, M. O., & Yan, J. (2026) · Journal of Agricultural Biological and Environmental Statistics · doi:10.1007/s13253-025-00720-7 · Cited by 6
“In the homoscedastic case, on σ we place a half-t prior with 3 degrees of freedom, denoted t + (3) (Gelman, 2006).”
Covariate‐based cepstral parameterizations for time‐varying spatial error covariances
Gladish, D. W., Wikle, C. K., & Holan, S. H. (2014) · Environmetrics · doi:10.1002/env.2259 · Cited by 3
“Alternatively, although not chosen here, a uniform prior for 2 (Gelman, 2006) or a normal prior for 0 (Holan and McElroy, 2012) could be chosen.”
BackgroundLeith, N. (2006) · Cited by 3
A Bayesian approach to estimating causal vaccine effects on binary post‐infection outcomes
Zhou, J., Chu, H., Hudgens, M. G., & Halloran, M. E. (2015) · Statistics in Medicine · doi:10.1002/sim.6573 · Cited by 4
Hierarchical Bayes Models for Variability
Kelly, D., & Smith, C. L. (2011) · Springer series in reliability engineering · doi:10.1007/978-1-84996-187-5_7 · Cited by 3
Bayesian Modeling of Follow-up Studies with Missing Data
Stamey, J. D., Bekele, B. N., & Powers, S. (2009) · Annals of Epidemiology · doi:10.1016/j.annepidem.2009.01.012 · Cited by 3
Lynch, M. L., & DeGruttola, V. (2014) · Journal of the Royal Statistical Society Series C (Applied Statistics) · doi:10.1111/rssc.12080 · Cited by 3
“…but yields a prior that is nearly improper and substantially concentrated near zero, which can cause difficulties in convergence (see, for instance, Gelman (2006) for a discussion within the context of hierarchical modelling); the hyperparameters (a, b) = (0.1, 0.1) that we employ reflect our goal…”
InfluentialParameter Expanded Algorithms for Bayesian Latent Variable Modeling of Genetic Pleiotropy Data
Xu, L., Craiu, R. V., Sun, L., & Paterson, A. D. (2014) · Journal of Computational and Graphical Statistics · doi:10.1080/10618600.2014.988337 · Cited by 3
“1), we follow the procedure in Gelman and Meng (1998) and use 30 grid points ranging evenly from 0 to 1 so that s(0) = 0 < s(1) < s(2) < .”
BackgroundMethodInfluentialHuang, Y., & Dagne, G. A. (2012) · Communications in Statistics - Simulation and Computation · doi:10.1080/03610918.2012.664229 · Cited by 3
“…component of the population parameter vector β; (ii) for the scale parameters σ 2 and 2, we assume a limiting noninformative IG prior distribution (Gelman, 2006) IG(0.01, 0.01) so that the distribution has mean 1 and variance 100; (iii) the prior for the variance-covariance matrix of the…”
BackgroundInfluentialSiddique, J., Crespi, C. M., Gibbons, R. D., & Green, B. L. (2010) · Statistics in Medicine · doi:10.1002/sim.4109 · Cited by 4
Clinically relevant graphical predictions from Bayesian joint longitudinal-survival models
Hatfield, L. A., & Carlin, B. P. (2012) · Health Services and Outcomes Research Methodology · doi:10.1007/s10742-012-0087-9 · Cited by 3
“The three variance parameters have support ½0;1Þ; so we choose half-Cauchy priors on the standard deviations, following Gelman (2006).”
MethodHierarchical Bayes estimation in small area estimation using cross-sectional and time-series data
Torabi, M., & Shokoohi, F. (2012) · Journal of Statistical Computation and Simulation · doi:10.1080/00949655.2012.721365 · Cited by 3
Breeding system and reproductive skew in a highly polygynous ant population
Haag-Liautard, C., Pedersen, J. S., Ovaskainen, O., et al. (2008) · Insectes Sociaux · doi:10.1007/s00040-008-1010-9 · Cited by 3
Logistic regression in sealed-bid auctions with multiple rounds: Application in Korean court auction
Kim, C., & Chang, W. (2010) · Expert Systems with Applications · doi:10.1016/j.eswa.2010.08.102 · Cited by 3
Malec, D. J., & Maples, J. J. (2008) · Statistics in Medicine · doi:10.1002/sim.3254 · Cited by 3
Cowles, M. K. (2012) · Springer texts in statistics · doi:10.1007/978-1-4614-5696-4_1 · Cited by 3
Webb, J. A., King, E. L., Reynoldson, T. B., & Padgham, M. (2014) · Freshwater Science · doi:10.1086/678949 · Cited by 3
“…is recommended for between-group variances in hierarchical models when there are few groups because other commonly used minimally informative priors (e.g., the inverse γ on variance, or uniform on standard deviation) can have substantial effects on the results in such circumstances (Gelman 2006).”
BackgroundMethodInfluentialLindhagen, L., Darkahi, B., Sandblom, G., & Berglund, L. (2014) · Statistics in Medicine · doi:10.1002/sim.5677 · Cited by 3
Imprecise structural reliability analysis using PC-Kriging
Schöbi, R., & Sudret, B. (2015) · Repository for Publications and Research Data (ETH Zurich) · doi:10.3929/ethz-a-010494143 · Cited by 3
“A wide variety of approaches have been proposed in the literature, such as Dempster-Shafer’s theory (Dempster 1967, Shafer 1976), probability-boxes (Ferson and Ginzburg 1996) and Bayesian hierarchical models (Gelman 2006).”
BackgroundTian, T., McLachlan, G. J., Dieters, M. J., et al. (2015) · PLoS ONE · doi:10.1371/journal.pone.0144370 · Cited by 3
“We followed Gelman [18] in assuming that Vβ = τ2I, and the hyperparameters β0, ν0, S0 and τ 2 were fixed and known.”
MethodElster, C., & Wübbeler, G. (2016) · Computational Statistics · doi:10.1007/s00180-015-0641-3 · Cited by 3
Modeling agreement on categorical scales in the presence of random scorers
Vanbelle, S., & Lesaffre, E. M. E. H. (2015) · Biostatistics · doi:10.1093/biostatistics/kxv036 · Cited by 3
Simultaneous Source Detection and Analysis Using a Zero-inflated Count Rate Model
Klumpp, J. A., & Brandl, A. (2015) · Health Physics · doi:10.1097/hp.0000000000000291 · Cited by 3
“This technique may work particularly well if a “halfCauchy” prior (that is, a Student’s t distribution with one degree of freedom truncated at zero) is put on λS before the initial sample measurement (Gelman 2006).”
MethodAppraising Between-Study Homogeneity, Small-Study Effects, Moderators, and Confounders
Veroniki, A. A., Huedo-Medina, T. B., & Fountoulakis, K. N. (2016) · Springer eBooks · doi:10.1007/978-3-319-25655-9_12 · Cited by 3
A novel method for quantifying overdispersion in count data and its application to farmland birds
McMahon, B. J., Purvis, G., Sheridan, H., SIRIWARDENA, G. M., & Parnell, A. C. (2016) · Ibis · doi:10.1111/ibi.12450 · Cited by 3
“For the random effect variance parameters, we used half-Cauchy t1ð0; 10Þ distributions, which are only weakly informative (Gelman 2006).”
MethodContributed Discussion on Article by Finegold and Drton
authors, V. (2014) · Bayesian Analysis · doi:10.1214/13-ba856d · Cited by 3
Estimating extinction risk with minimal data
Weiss‐Lehman, C., Davies, K. F., Clements, C. F., & Melbourne, B. A. (2017) · Biological Conservation · doi:10.1016/j.biocon.2017.06.020 · Cited by 5
Collis, J., Hill, M. R. S., Nicol, J. R., et al. (2017) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2017.08.015 · Cited by 3
Phase 3 adaptive trial design options in treatment of complicated urinary tract infection
Viele, K., Mundy, L. M., Noble, R. B., Li, G., Broglio, K. R., & Wetherington, J. D. (2018) · Pharmaceutical Statistics · doi:10.1002/pst.1892 · Cited by 4
Yan, Z., Xu, J., & Ruan, X. (2019) · Hydrology Research · doi:10.2166/nh.2019.076 · Cited by 2
Isogawa, N., Takeda, K., Maruo, K., & Daimon, T. (2019) · Therapeutic Innovation & Regulatory Science · doi:10.1177/2168479019862531 · Cited by 3
Bayesian Analysis of the Relationship Between Belief Conflict and Occupational Dysfunction
Kyougoku, M., & Teraoka, M. (2019) · American Journal of Occupational Therapy · doi:10.5014/ajot.2019.027615 · Cited by 5
Clustering ultrasonic waves propagation time: A hierarchical polynomial semiparametric approach
Zuanetti, D. A., da Paz, R. F., Rodrigues, T., & Mesquita, E. (2021) · Applied Stochastic Models in Business and Industry · doi:10.1002/asmb.2609 · Cited by 2
Doll, J. C., & Jacquemin, S. (2021) · PeerJ · doi:10.7717/peerj.11032 · Cited by 3
An Objective Prior from a Scoring Rule
Walker, S. G., & Villa, C. (2021) · Entropy · doi:10.3390/e23070833 · Cited by 4
A Bayesian Approach to Envelope Quantile Regression
Lee, M., Chakraborty, S., & Su, Z. (2021) · Statistica Sinica · doi:10.5705/ss.202020.0109 · Cited by 3
“More specifically, we consider the following: (i) a = b = 10−4 in the prior π(σ) [16]; (ii) each element of e in π(η | σ,A) is 10−3, and M in π(η | σ,A) is 10−6 times a diagonal matrix with diagonal elements generated from a χ2 distribution with degrees of freedom one; (iii) A0 in π(A) is the zero matrix; (iv) the cov…”
BackgroundCan a rational expectation storage model explain the USDA ending grain stocks forecast errors?
Zhang, T., & Li, Z. (2021) · Journal of Futures Markets · doi:10.1002/fut.22282 · Cited by 3
Multi‐scale patterns in the occurrence of an ephemeral pool‐breeding amphibian
Halstead, B. J., Rose, J. P., Clark, D. R., Kleeman, P. M., & Fisher, R. N. (2022) · Ecosphere · doi:10.1002/ecs2.3960 · Cited by 3
Interval Finite Element Method
Nayak, S., & Chakraverty, S. (2018) · Elsevier eBooks · doi:10.1016/b978-0-12-812973-9.00002-3 · Cited by 3
Modeling spatio-temporal activity dynamics of the small relict marsupial Dromiciops gliroides
Vazquez, M. S., Schenone, L., Rodríguez‐Cabal, M. A., et al. (2022) · Mammalian Biology · doi:10.1007/s42991-022-00331-6 · Cited by 3
“We choose non-informative priors for all co-variate coefficients (normal distributions with mean = 0 and variance = 5) and half-Cauchy distributions for the variance component in the random hierarchical levels (Gelman 2006).”
BackgroundKim, B., Niu, X. M., Hunter, D. R., et al. (2023) · The Annals of Applied Statistics · doi:10.1214/23-aoas1762 · Cited by 3
Meta-analytic framework for modeling genetic coexpression dynamics
Kinzy, T. G., Starr, T. K., Tseng, G. C., & Ho, Y.-Y. (2019) · Statistical Applications in Genetics and Molecular Biology · doi:10.1515/sagmb-2017-0052 · Cited by 3
Modeling Crop Phenology in the US Corn Belt Using Spatially Referenced SMOS Satellite Data
Lewis‐Beck, C., Zhu, Z., Walker, V. A., et al. (2020) · Journal of Agricultural Biological and Environmental Statistics · doi:10.1007/s13253-020-00419-x · Cited by 3
Markov Interacting Importance SamplersPreprint
Mendes, E. F., Scharth, M., & Kohn, R. (2015) · SSRN Electronic Journal · doi:10.2139/ssrn.2569488 · Cited by 3
A Comparison of Adjusted Bayes Estimators of use in Small Area Estimation
Fabrizi, E., Ferrante, M. R., & Pacei, S. (2007) · Aisberg (University of Bergamo) · Cited by 3
Flexible Bent-Cable Models for Mixture Longitudinal Data
Khan, S. A. (2010) · UWSpace (University of Waterloo) · Cited by 3
Prior Selection for Panel Vector AutoregressionsPreprint
Korobilis, D. (2015) · SSRN Electronic Journal · doi:10.2139/ssrn.2602508 · Cited by 3
Passive Acoustics: A Multifaceted Tool for Marine Mammal Conservation
Phillips, G. T. (2016) · DukeSpace (Duke University) · Cited by 3
Jiang, L., Li, L., & Yao, W. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1607.00098 · Cited by 3
Hierarchical Bayesian models for estimating the extent of plant pest invasions
Stanaway, M. (2011) · Cited by 3
Hodges, J. S. (2008) · Journal of the American Statistical Association · doi:10.1198/016214507000000644 · Cited by 3
Selection of lamb size and early pregnancy in Soay sheep ( Ovies aries )Preprint
Janeiro, M. J., Henshaw, J. M., Pemberton, J. M., et al. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.09.16.299685 · Cited by 3
Parsons, J., Niu, X. M., & Bao, L. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.05346 · Cited by 3
Bayesian Modeling of Spatial Molecular Profiling Data via Gaussian ProcessPreprint
Li, Q., Zhang, M., Xie, Y., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2012.03326 · Cited by 3
Lin, W., Donohue, M., Insel, P. S., Schwartzman, A., & Thompson, W. K. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.03.10.434854 · Cited by 3
Liew, S. X., Afrasiabi, M., & Austerweil, J. L. (2019) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/aes5f · Cited by 3
X vs. Y: An Analysis of Intergenerational Differences in Transport Mode Use Among Young AdultsPreprint
Krueger, R., Rashidi, T. H., & Vij, A. (2018) · SSRN Electronic Journal · doi:10.2139/ssrn.3224326 · Cited by 3
Ren, W., Yu, B., Chen, Y., et al. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.02422 · Cited by 3
“A well-known computing approach for Bayesian inference, the Markov chain Monte Carlo (MCMC) method (Gelman, 2006), was used in this study to better approximate the target posterior distribution.”
MethodDoors and corners of variance partitioning in statistical ecologyPreprint
Schulz, T., Saastamoinen, M., & Vanhatalo, J. P. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.10.17.464682 · Cited by 3
“Superpopulation variance partitioning requires a more carefully thought out model structure and is more sensitive to choice of priors (Gelman 2005, 2006).”
BackgroundInvestigating the role of dispersal on the genetic structure of wild populations
Powell, J. H. (2014) · Montana State University ScholarWorks (Montana State University) · Cited by 3
SPDE based modeling of large space-time data setsPreprint
Sigrist, F., & Stahel, W. A. (2014) · arXiv (Cornell University) · Cited by 3
Zhou, G. C., Song, S., & Szczesniak, R. D. (2023) · Statistics in Medicine · doi:10.1002/sim.9758 · Cited by 2
Tamai, Y., Noda, A., & Yamamoto, E. (2023) · Journal of Chromatography A · doi:10.1016/j.chroma.2023.464136 · Cited by 2
Modeling basal body temperature data using horseshoe process regression
Chase, E. C., Taylor, J. M. G., & Boonstra, P. S. (2023) · Statistics in Medicine · doi:10.1002/sim.9991 · Cited by 2
A basket trial design based on constrained hierarchical Bayesian model for latent subgroups
Takeda, K., Hashimoto, A., Liu, S., et al. (2024) · Journal of Biopharmaceutical Statistics · doi:10.1080/10543406.2024.2311851 · Cited by 2
“Referring to Gelman (2006), the hyperparameters μ and τ followed a normal distribution μ , N ð(cid:0) 2 : 20 ; 100 Þ and a uniform distribution τ , Uniform ð 0 ; 1 : 5 Þ , respectively.”
Beyer, S. G., Sogard, S. M., Stafford, D. M., Kashef, N. S., Rodríguez, A., Alonzo, S. H., & Field, J. C. (2024) · Canadian Journal of Fisheries and Aquatic Sciences · doi:10.1139/cjfas-2023-0253 · Cited by 6
Chinembiri, T. S., Mutanga, O., & Dube, T. J. (2024) · Global Environmental Change Advances · doi:10.1016/j.gecadv.2024.100010 · Cited by 2
Leveraging single-case results to Bayesian hierarchical modelling
Si, S., Gu, J., & Tian, M. (2024) · Computational Statistics · doi:10.1007/s00180-024-01516-y · Cited by 2
“For the Bayesian hierarchical model, we assume that the batting averages i , i = 1, … , 18 follow a common normal distribution, i.e., where the prior p H ( 𝛾 , 2 ) is weakly informative and recommended by Gelman (2006).”
Aslanyan, V., Pickering, T. A., Nuño, M. M., Renfro, L. A., Pa, J., & Mack, W. J. (2024) · Pharmaceutical Statistics · doi:10.1002/pst.2443 · Cited by 2
A Phase I Dose‐Finding Design Incorporating Intra‐Patient Dose Escalation
Guo, B., & Liu, S. (2024) · Pharmaceutical Statistics · doi:10.1002/pst.2461 · Cited by 2
Hotter temperatures alter riparian plant outcomes under regulated river conditions
Palmquist, E. C., Ogle, K., Butterfield, B. J., et al. (2025) · Ecological Monographs · doi:10.1002/ecm.1645 · Cited by 2
Alvares, D., Barrett, J., Mercier, F., Roumpanis, S., Yiu, S., Castro, F., Schulze, J., & Zhu, Y. (2025) · Statistics in Medicine · doi:10.1002/sim.10322 · Cited by 5
Bayesian Variance Change Point Detection With Credible Sets
Cappello, L., & Padilla, O. H. M. (2025) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2025.3548012 · Cited by 14
“Comparing PRISCA to PRISCA-Half-Cauchy ( ξ = 25 as in [55]) suggests that, in this specific setting, the prior on the variance does not seem to have an effect.”
Influentialvan de Wiel, M. A., Leday, G. G. R., Heymans, M. W., van Zwet, E. W., Zwinderman, A. H., & Hoogland, J. (2025) · Statistical Methods in Medical Research · doi:10.1177/09622802251338440 · Cited by 11
Lin, Y.-S., Hsieh, N., Keshava, N., Chou, W., Peterson, G. C. L., & Gift, J. S. (2025) · Environmental Science & Technology · doi:10.1021/acs.est.5c00627 · Cited by 2
Assessing Occurrence Patterns of Shallow Hikurangi Slow Slip Events Using Renewal Processes
Perez‐Silva, A., Wang, T., Wallace, L. M., et al. (2025) · Geophysical Research Letters · doi:10.1029/2025gl116605 · Cited by 2
Melo-Vega-Angeles, O., & Chuquillanqui-Lichardo, B. (2025) · Finance research letters · doi:10.1016/j.frl.2025.108131 · Cited by 3
Being Bayesian about learning Bayesian networks from hybrid data
Grzegorczyk, M. A. (2025) · International Journal of Approximate Reasoning · doi:10.1016/j.ijar.2025.109549 · Cited by 6
Parvej, M., Devashish, D., Ashraf, F., & Khan, A. A. (2025) · International Journal of Applied and Computational Mathematics · doi:10.1007/s40819-025-02031-4 · Cited by 3
Shen, J.-W., Moodie, E., & Golchi, S. (2025) · Statistical Methods in Medical Research · doi:10.1177/09622802251387430 · Cited by 4
Almeida, J. C., Costa, A. B., Özkan, B., et al. (2025) · Scientific Reports · doi:10.1038/s41598-025-22781-x · Cited by 2
Exploration of virtual control groups and Bayesian approaches for rat fertility studies
Bowman, C., Li, D., Tran, T. D. B., et al. (2025) · Regulatory Toxicology and Pharmacology · doi:10.1016/j.yrtph.2025.106004 · Cited by 2
From tetrachoric to kappa: How to assess reliability on binary scales
Vanbelle, S. (2025) · British Journal of Mathematical and Statistical Psychology · doi:10.1111/bmsp.70021 · Cited by 4
An analytical study of worker well-being and COVID-19 impact using Bayesian panel modeling
Nakakita, M., Toyabe, T., Kubota, N., Saito, W., & Nakatsuma, T. (2025) · Healthcare Analytics · doi:10.1016/j.health.2025.100434 · Cited by 3
Enhancing risk prediction with Bayesian hierarchical frailty models in Stan
Parvej, M., Khan, N., Ashraf, F., & khan, A. A. (2026) · Communication in Statistics- Theory and Methods · doi:10.1080/03610926.2025.2603651 · Cited by 2
Metadata
- Authors
- Gelman, A.
- Affiliations
- Columbia University (US)
- Corresponding author
- Gelman, Andrew
- ORCID
- Gelman, Andrew
- Venue
- Bayesian Analysis
- Publisher
- Institute of Mathematical Statistics
- Published
- 2006-09-01
- Volume
- 1
- Issue
- 3
- Pages
- 515-534
- ISSN
- 1931-6690 (ISSN-L)
- 1936-0975
- OpenAlex
- W1536497620
- Semantic Scholar
- 63d40805d4635b70f6b5aa07fc2c9821480cfc1e
- Cited by
- 4,050
- References
- 34
- Citations by year
- 2026: 194 · 2025: 218 · 2024: 237 · 2023: 237 · 2022: 282 · 2021: 312 · 2020: 320 · 2019: 281 · 2018: 266 · 2017: 263 · 2016: 235 · 2015: 217 · 2014: 176 · 2013: 142 · 2012: 184
- FWCI
- 78.70
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- cc-by
- Locations
- Bayesian Analysis (publishedVersion, cc-by)
- Project Euclid (Cornell University) (submittedVersion)
- Funders
- National Science Foundation
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Distribution Estimation and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Methods and Mixture Models
- Keywords
- prior distributions, hierarchical models, scale parameter, non-informative priors, weakly informative priors, uniform prior, analysis of variance
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-02, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-02 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.