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 3,501–3,600 of 4,089
An Introduction to Inductive Statistical Inference: from Parameter Estimation to Decision-MakingPreprint
van Elst, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.10173 · Cited by 3
“Gelman (2006) [33], Gill (2015) [39, p 178], Kruschke (2015) [58, p 558], and McElreath (2016) [68, pp 260, 363].”
BackgroundComparison of Heterogeneity and Heterogeneity Interval Estimators in Random-Effects Meta-Analysis
Boedeker, P. (2018) · University of North Texas Digital Library (University of North Texas) · doi:10.12794/metadc1157553
Hierarchical model for forecasting the outcomes of binary referenda
Wiśniowski, A., Bijak, J., Forster, J. J., et al. (2018) · Computational Statistics & Data Analysis · doi:10.1016/j.csda.2018.09.007
“Hence, following recommendations of Gelman (2006), we investigated (i) diffuse half-normal prior distributions for precisions, and (ii) uniform distributions over large range for standard deviations.”
MethodGhosal, S. (2018) · ThinkIR: The University of Louisville's Institutional Repository (University of Louisville) · doi:10.18297/etd/3039
“1, which is common choice to induce some shrinkage of the regression coe cients in hierarchical models (Gelman, 2006).”
BackgroundShantz, K. (2018) · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign) · Cited by 1
A Model for Understanding Recognition Validity
Lee, M. D., Doering, S., & Carr, A. (2019) · Computational Brain & Behavior · doi:10.1007/s42113-019-0023-y
“The standard deviations measuring the extent of the noise in sending and receiving are given flat priors over a large plausible range of values (Gelman, 2006).”
BackgroundTrainor, P. J. (2018) · ThinkIR: The University of Louisville's Institutional Repository (University of Louisville) · doi:10.18297/etd/3023
Characterizing and responding to uncertainty in climate change
Lemoine, D. M. (2011) · eScholarship (California Digital Library)
Using Bayesian Multilevel Models to Control for Multiplicity among Means
Zweifel, M. J. (2018) · Lincoln (University of Nebraska)
An Adaptable Long-term Condition Workload Prediction Model for Primary Health Care
Devananda, M. (2018) · Otago University Research Archive (University of Otago)
Robinson, A. P., Ryu, D., & Lee, H. J. (2019) · Phytopathology · doi:10.1094/phyto-05-18-0168-r
Montoya, I., Aldasoro, E., Calvo, M., Audicana, C., Esnaola, S., & Grande, A. G. (2019) · Gaceta Sanitaria · doi:10.1016/j.gaceta.2018.12.004
Lucio, E. (2018) · The University of Queensland · doi:10.14264/uql.2018.713
Strategic Leniency by the Court of Justice of the European Union: A Choice for Integration?Preprint
Castro-Montero, J. L. (2017) · SSRN Electronic Journal · doi:10.2139/ssrn.2978615
Integrating subgroups with mixed-type endpoints in early phase oncology trials
Zhao, L., & Koschmann, C. J. (2019) · Statistical Methods in Medical Research · doi:10.1177/0962280219840050
Control of Residential Space Heating for Demand Response Using Grey-box Models
Hedegaard, R. E. (2018) · doi:10.7146/aul.317.215
Jang, E. J., & Kim, D. H. (2019) · Communications for Statistical Applications and Methods · doi:10.29220/csam.2019.26.2.205
“The hyperprior for µ is assumed to be Cauchy prior (Gelman et al., 2008) and the hyperprior for σ2 is assumed as shrinkage prior to avoid difficulties associated with improper priors of the form π(σ2) ∝ 1/σ2 (Gelman, 2006; Nandram et al., 2013) as: π(µ) = 1 π ( 1 + µ2 ) , π (σ2) = 1( 1 + σ2 )2 .”
BackgroundParameter clustering in Bayesian functional PCA of fMRI dataPreprint
Margaritella, N., Inácio, V., & King, R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1904.11758
“However, alternative prior distributions for the precision parameter have been proposed (Gelman, 2006).”
BackgroundSpatial Data and Spatial Statistics
Haining, R., & Li, G. (2019) · Springer eBooks · doi:10.1007/978-3-642-36203-3_71-1
Quantifying evolution in wild populations
Ramakers, J. J. C. (2019) · doi:10.18174/472942 · Cited by 1
Bayesian adaptive methods to incorporate preclinical data into phase I clinical trials
Zheng, H. (2019) · Lancaster EPrints (Lancaster University) · doi:10.17635/lancaster/thesis/573
Bayesian forecasting of electoral outcomes with new parties' competitionPreprint
Montalvo, J. G., Papaspiliopoulos, O., & Stumpf-Fétizon, T. (2016) · RePEc: Research Papers in Economics · doi:10.48550/arxiv.1612.03073
McParland, D., Baron, S., O'Rourke, S., et al. (2016) · Maynooth University ePrints and eTheses Archive (Maynooth University) · doi:10.48550/arxiv.1608.03091
“Informative prior distributions are to be preferred (such as the GP prior in the second level), but hyper-parameters for GP models are notoriously difficult to obtain so we follow the weakly-informative convention of Gelman (2006).”
BackgroundAn adaptive spatio-temporal smoothing model for estimating trends and step changes in disease risk
Rushworth, A., Lee, D., & Sarran, C. E. (2014) · Strathprints: The University of Strathclyde institutional repository (University of Strathclyde) · doi:10.48550/arxiv.1411.0924
“Our choice of inverse gamma priors here is due to their conjugacy, but alternatives could be explored such as the half-Cauchy distribution as suggested by Gelman (2006).”
BackgroundEstimating global exposure effect from longitudinal data on multiple surrogates
Bandyopadhyay, S., Ganguli, B., & Chatterjee, A. (2019) · Communications in Statistics Case Studies Data Analysis and Applications · doi:10.1080/23737484.2019.1631722
Action Intentions and the Puzzle of Non-Willed Behaviour
Colton, J. (2019) · PEARL (University of Plymouth) · doi:10.24382/814
A Bayesian Constancy-Enforced Non-Inferiority Design in Medical Device Trials with a Binary Endpoint
Yang, Y., Xu, Y., Lu, N. T., & Tiwari, R. C. (2019) · ICSA book series in statistics · doi:10.1007/978-3-030-15310-6_13
Ong, V. M.-H., Nott, D. J., Choi, T., et al. (2019) · Foundations of Data Science · doi:10.3934/fods.2019006
Bayesian methods for time series of count data
Obeidat, M., Liu, J., Osgood, N. D., et al. (2019) · Communications in Statistics - Simulation and Computation · doi:10.1080/03610918.2019.1655574
“The full conditional posterior distribution of r2V is pðr2V jb, r, g, yÞ ¼ pðr2V jr, gÞ ¼ f ðg1, :::, gtj/, r2VÞpðr2VÞ / ðr2VÞ n 2 exp Xn t¼1ðgt l t Þ2=2Rt r2V ( ) pðr2VÞ pðr2VÞ Using the uniform prior for rV (Gelman 2006), the prior for r2V is pðr2VÞ / ðr2VÞ 1 2Ið0, 100Þ.”
A Semiparametric Stochastic Frontier Model with Correlated Effects
Hajargasht, G., & Griffiths, W. E. (2019) · Emerald Group Publishing Limited eBooks · doi:10.1108/s0731-90532019000040b002
A Multiscale Spatially Varying Coefficient Model for Regional Analysis of Topsoil Geochemistry
Kim, K., Kim, H., Kim, V., et al. (2019) · Journal of Agricultural Biological and Environmental Statistics · doi:10.1007/s13253-019-00379-x
Novel Bayesian methodology in multivariate problems.
Kundu, D. (2019) · doi:10.18297/etd/3293
Application of Hierarchical Bayesian Model in Ophtalmological Study
Tojtovska, B., Ribarski, P., & Ljubic, A. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-33110-8_10
Incorporating external evidence syntheses in the design and analysis of trials
Clayton, G. (2019) · Bristol Research (University of Bristol)
On the two-dataset problemPreprint
MacEachern, S. N., & Miyawaki, K. (2019) · arXiv (Cornell University)
Bayesian hierarchical modeling for the linkages between air pollution and population health
Hao, Y., Landrine, H., Ploeg, M. V., et al. (2019) · Elsevier eBooks · doi:10.1016/b978-0-12-815822-7.00004-2
Keizer, S., Zhan, Z. Z., Vasan, R. S., & van den Heuvel, E. R. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.06398
“r distribution for ˝ kis to constrain the intra-individual variance before and after treatment away from very large values in case of insucient number of follow-up measurements for some individuals (Gelman and others, 2006). Further reparametrization of ˝and Cholesky factorization of the correlation matrix are used on…”
MethodA unified view on Bayesian varying coefficient modelsPreprint
Franco‐Villoria, M., Ventrucci, M., & Rue, H. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1806.02084
Prediction in the Presence of Missing CovariatesPreprint
Page, G. L., Quintana, F. A., & Mueller, P. J. (2019) · arXiv (Cornell University)
Cavadino, A., Prieto‐Merino, D., & Morris, J. K. (2020) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.4948
“The minimally informative priors used throughout are described in Section S1; normal distributions were used for estimation of means (eg, average PRRs for each medication/CA or group of medications/CAs), and uniform distributions were used for variance parameters.(19) Minimally informative priors were used, as our mai…”
MethodNonstandard flexible regression via variational Bayes
Ormerod, J. T. (2020) · Elsevier eBooks · doi:10.1016/b978-0-12-815862-3.00010-x
Bayesian modelling for determining material properties
Kumar, A., Shrivastava, R., & Singh, K. H. (2018) · International Conference on Recent Innovations in Electrical, Electronics & Communication Engineering (ICRIEECE) · doi:10.1109/icrieece44171.2018.9009196
“But their usage has been severely discouraged by the works of [19] and [20].”
BackgroundMethodInfluentialAn incomplete taxonomy of Bayesian models with examples from industrial statistics applications
Hamada, M. S. (2020) · Quality Engineering · doi:10.1080/08982112.2020.1723626
Cao, Y., & Mukhopadhyay, N. D. (2020) · Sankhya B · doi:10.1007/s13571-019-00222-w
Bayesian Inferences on Uncertain Ranks and Orderings: Application to Ranking Players and Lineups
Barrientos, A. F., Sen, D., Page, G. L., & Dunson, D. B. (2022) · Bayesian Analysis · doi:10.1214/22-ba1324 · Cited by 7
“are less informative and induce shrinkage towards the mean , with prior 3 producing more shrinkage than prior 2. Prior 2 is motivated by Park and Casella (2008) and Hans (2009, 2010), and prior 3 by Gelman (2006) and Polson and Scott (2012). Since the idea is to simulate data that mimic the complexities of the NBA lin…”
BackgroundHierarchical Bayesian propulsion power models for marine vesselsPreprint
Solonen, A., Maraia, R., Springer, S., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.11267
“See [18] about setting priors for variance parameters in hierarchical models6.”
Sample bias related to household role
Hitczenko, M. (2021) · doi:10.29338/wp2021-09
Murayama, K., & Kawano, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2005.03419
“This hyperprior is related to a weakly informative prior which was proposed in the discussion of prior over variance in hierarchical model [19, 20] and a general definition was also given [21].”
BackgroundShrinkage estimation for dose-response modeling in phase II trials with multiple schedules
Günhan, B. K., Meyvisch, P., & Friede, T. (2020) · Statistics in Biopharmaceutical Research · doi:10.1080/19466315.2020.1850519
“This is similar to the meta-analysis of few studies, in which the estimation of the between-trial heterogeneity has gained considerable attention in the literature (Gelman 2006).”
ResultGhattas, A. E., Beichel, R., & Smith, B. J. (2020) · Statistical Methods in Medical Research · doi:10.1177/0962280220920894
Inferring the Allelic Series at QTL in Multiparental PopulationsPreprint
Crouse, W. L., Kelada, S. N. P., & Valdar, W. S. J. (2020) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2020.05.23.112326
“The half-t is a preferred prior choice for variance components in normally-distributed models due to its behavior at the boundary (ϕ = 0) and its convenient representation as the product of two conditionally-conjugate latent variables (Gelman 2006).”
BackgroundBayesian Modelling to Assist Inference on Health Outcomes in Occupational Health Surveillance
Tierney, N. J., Clifford, S., Drovandi, C., & Mengersen, K. L. (2020) · Lecture notes in mathematics · doi:10.1007/978-3-030-42553-1_13
Adaptive Bayesian Nonparametric Smoothing with Markov Random Fields and Shrinkage Priors
Faulkner, J. R. (2019) · ResearchWorks at the University of Washington (University of Washington) · Cited by 2
Grunwald, G. K., Arnett, J. A., Liu, W., & Ho, P. M. (2020) · Biometrical Journal · doi:10.1002/bimj.201900148
“There has been much discussion about priors for variance parameters in hierarchical models, for example, Gelman (2006), and much subsequent work, but substantially less about priors for covariance matrices in hierarchical models.”
BackgroundLearning Hamiltonian Monte Carlo in RPreprint
Thomas, S., & Tu, W. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.16194
“R> plot(fm1_hmc, burnin=200) Note that the Inverse Gamma distribution is not always an optimal prior when the support is strictly positive.”
MethodInfluentialBayesian Deconvolution of Rate Pressure Data Using Regularizing Priors
Ravikumar, A., & Lee, J. (2020) · SPE Latin American and Caribbean Petroleum Engineering Conference · doi:10.2118/198960-ms
Quantification of Uncertainty in Model Based Type Well Construction
Ravikumar, A., & Lee, J. R. (2020) · SPE Latin American and Caribbean Petroleum Engineering Conference · doi:10.2118/199118-ms
Constructing an Informative Prior Distribution of Noises in Seasonal Adjustment
Guo, L. (2020) · uO Research (University of Ottawa) · doi:10.20381/ruor-25293
Bayesian inference under small sample size -- A noninformative prior approachPreprint
He, J., & Guan, X. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.06110
“Gelman [29] discussed the inverse gamma distribution IG(ǫ, ǫ) as an attempt at noninformativeness within the conjugate family, with ǫ set to a low value such as 1 or 0.”
BackgroundInfluentialA new mixed‐effects regression model for the analysis of zero‐modified hierarchical count data
Bertoli, W., Conceição, K. S., Andrade, M. G., et al. (2020) · Biometrical Journal · doi:10.1002/bimj.202000046
“The distribution was cited by Gelman (2006) and Polson and Scott (2012) as an alternative to the traditionally used Inverse Gamma prior.”
MethodMarshall, C. (2020) · arXiv (Cornell University)
Part Performance Measurement, Analysis and Optimization for Binder Jetting Additive Manufacturing
Wang, M. (2020) · UWSpace (University of Waterloo) · Cited by 1
Empirical estimates for heteroscedastic hierarchical dynamic normal models
Ghoreishi, S. K., & Wu, J. (2020) · Journal of the Korean Statistical Society · doi:10.1007/s42952-020-00093-2
Ogero, M. O., Orwa, J., Odhiambo, R., et al. (2020) · Research Square · doi:10.21203/rs.3.rs-122540/v1
Multilevel Item Response Model with Heterogeneous Within-Cluster Variances
Kara, Y., & Kamata, A. (2020) · The Journal of Experimental Education · doi:10.1080/00220973.2020.1858015
“Priors such as the half-Cauchy (Gelman, 2006) and truncated shifted inverse-gamma priors (Fox et al., 2017; Klotzke & Fox, 2019a, 2019b) can be good candidates.”
BackgroundBayesian methods for borrowing information in clinical drug development
Günhan, B. K. (2020) · eDiss (Georg-August-Universität Göttingen) · doi:10.53846/goediss-8220
Bayesian inference with tmbstan for a state-space model with VAR(1) state equationPreprint
Cao, Y., & Tufto, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.05635
Bayesian Latent Variable Models for the Analysis of Experimental Psychology DataPreprint
Merkle, E. C., & Wang, T. (2014) · SSRN Electronic Journal · doi:10.2139/ssrn.2536989
Estimación de áreas pequeñas utilizando imputación múltiple en modelos logísticos de tres parámetros
Piñerez, T., & Fernando, C. (2020)
Bayesian estimation of the Eurozone currency union effect
Carpio, R., & Guo, M. (2021) · Review of International Economics · doi:10.1111/roie.12518
Atlas, W. I., Holt, C. A., Selbie, D. T., Connors, B. M., Cox‐Rogers, S., Carr‐Harris, C., Hertz, E., & Moore, J. W. (2021) · bioRxiv (Cold Spring Harbor Laboratory) · doi:10.1101/2021.03.22.436497
Skewed Spatial Modeling for Arsenic Contamination in Bangladesh
Zhang, Q. (2019) · Spectrum Research Repository (Concordia University)
Czahor, M. S., & Meeker, W. Q. (2017) · Annual Conference of the PHM Society · doi:10.36001/phmconf.2017.v9i1.2482
“For more information on prior distributions for variance parameters, see Gelman (2006).”
BackgroundBayesian Functional Principal Components Analysis via Variational Message PassingPreprint
Nolan, T. H., Goldsmith, J., & Ruppert, D. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.00645
“This auxiliary variable-based hierarchical construction facilitates arbitrarily non-informative priors on standard deviation parameters (Gelman, 2006).”
MethodBayesian Hierarchical Methods for Network Meta-Analysis
Zhang, C., & Yuan, Z. (2014)
Modeling a sequence of multinomial data with randomly varying probabilitiesPreprint
Deb, S., Roy, R., & Das, S. (2021) · arXiv (Cornell University)
Ecology of Vector-borne Diseases at the Interface of Theory and Data
Kamiya, T. (2020) · TSpace (University of Toronto)
Angelini, J., Bortolotto, E. B., Faviere, G. S., et al. (2021) · Research Square · doi:10.21203/rs.3.rs-377627/v1
Multilevel Dynamic Twin Modeling PreprintPreprint
Schuurman, N. K., Zheng, Y., & Dolan, C. V. (2020) · doi:10.31234/osf.io/x52af
Measuring the Trend Real Interest Rate in a Data-Rich EnvironmentPreprint
Fu, B. (2020) · SSRN Electronic Journal · doi:10.2139/ssrn.3539046
Variational Bayesian Learning of SMoGs: Modelling and Their Application to Synthetic Aperture Radar
Roussos, E. (2021) · Mathematical and Computational Applications · doi:10.3390/mca26020045
Fox, J., & Smink, W. A. C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.10107
Mitsunari, Y., & Yoshino, A. (2021) · Agricultural Information Research · doi:10.3173/air.30.45
Augusteijn, H. E. M., van Aert, R. C. M., & van Assen, M. A. L. M. (2021) · OSF Preprints (OSF Preprints) · doi:10.31219/osf.io/avkgj
“The use of a uniform prior can be a good starting point, but puts relatively much weight on unreasonably large values (Gelman, 2006; Williams et al., 2018).”
BackgroundMethodInfluentialDissecting the interaction of nitrogen, density, and genetics in maize
King, K. A. (2021) · Iowa State University Digital Repository (Iowa State University) · doi:10.31274/etd-20210609-91 · Cited by 1
“…d1h, d2h, dnh, d2nh, dn2h, d2n2h and the blocking term ijk had priors of the following form, in which h represents any one of the hyper parameters and HC represents a half-Cauchy distribution for the variance as described in (Gelman, 2006): n ~ N(,2), ~ N(0,10000), 2 ~ HC(36.1).”
MethodInfluentialCollective risk assessment in Affordable Care Act markets: A Bayesian hierarchical modelPreprint
de Oyarbide, J. I., & Paulo, R. (2020) · SSRN Electronic Journal · doi:10.2139/ssrn.3659528
Estimating Market Fundamentals from REIT dataPreprint
Geltner, D., Kumar, A., & de Minne, A. V. (2021) · SSRN Electronic Journal · doi:10.2139/ssrn.3881407
Bayesian Statistical Modeling of Metagenomics Sequencing Data
Jiang, S. (2021) · SMU Scholar (Southern Methodist University)
Externalities and Redistribution
Rueda, D., & Stegmueller, D. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108681339.004
Castel, C., Sommen, C., Chatignoux, E., Strat, Y. L., & Alioum, A. (2021) · The International Journal of Biostatistics · doi:10.1515/ijb-2021-0030
Hamiltonian Monte Carlo with Random Effect for Analyzing Cyclist Crash Severity
Rezapour, M., & Ksaibati, K. (2021) · Signals · doi:10.3390/signals2030032
“where L is an identity matrix of In, D1/2 is a matrix with diagonal values of λ, where λ follows half-t student distribution [20].”
BackgroundIncome, Income Expectations, Redistribution, and Insurance
Rueda, D., & Stegmueller, D. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108681339.002
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.