The Savage Award

The Savage Award, named in honor of Leonard J. "Jimmie" Savage, is bestowed each year to two outstanding doctoral dissertations in Bayesian econometrics and statistics, one each in:

  • Theory and Methods:   for a dissertation that makes important original contributions to the foundations, theoretical developments, and/or general methodology of Bayesian analysis.
  • Applied Methodology:   for a dissertation that makes outstanding contributions with novel Bayesian analysis of a substantive problem that has potential to impact statistical practice in a field of application.

Each award is accompanied by a monetary prize.

The award was instituted by the NBER-NSF Seminar in Bayesian Inference in Econometrics and Statistics in 1977 with an endowed fund supported by royalties from a series of books authored and edited under the auspices of  the Seminar on Bayesian Inference in Econometrics. ISBA and the ASA Section on Bayesian Statistical Science (SBSS) joined as co-sponsors in 1993.   Additional contributions to the endowment include royalties from the Handbook of Bayesian Econometrics edited by John Geweke, Gary Koop and Herman van Dijk. and donations by past recipients of the Savage Prize and members of ISBA.

Leonard J. "Jimmie" Savage

Born 20 November 1917, Jimmie Savage was graduated from the University of Michigan and later worked at the Institute for Advanced Study in Princeton, New Jersey, the University of Chicago, and the Statistical Research Group at Columbia University. Though his thesis advisor was Sumner Myers, he also credited Milton Friedman and W. Allen Wallis as his statistical mentors.

His most noted work was the 1954 book Foundations of Statistics, in which he put forward a theory of subjective and personal probability and statistics which forms one of the strands underlying Bayesian statistics and has applications to game theory.

One of Savage's indirect contributions was his discovery of the work of Louis Bachelier on stochastic models for asset prices and the mathematical theory of option pricing. Savage brought the work of Bachelier to the attention of Paul Samuelson. It was from Samuelson's subsequent writing that random walk (and subsequently Brownian motion) became fundamental to mathematical finance.

In 1951 he introduced the Minimax regret criterion used in decision theory. The Hewitt-Savage zero-one law is (in part) named after him.

Eligibility and Application Procedure

The Bylaws specify that all Ph.D. theses that have not been submitted in a previous year are eligible; thus a dissertation may be nominated only once. The Prize Committee interprets the phrase "Ph.D. Thesis" to mean a dissertation in final form: approved by the student's committee or examining board, for example (final University approval is not required). A dissertation may be nominated by the author, by the advisor, the department chair, or by any ISBA or SBSS member (joining ISBA is easy)  Nomination is made by on-line electronic submission of the dissertation along with a letter that describes the main theoretical, methodological, and/or applied contributions of the thesis and specifies for which award the thesis is being nominated --- either Theory and Methods or Applied Methodology. Nominating letters should be written in English, and it is recommended that dissertations also be written in English (the Savage Award Committee may require an English translation for full consideration of theses written in other languages). Dissertations and letters must be submitted electronically in pdf format. Nominations may be submitted at any time, but must be submitted by 31 May  (midnight UTC/GMT, 8pm EDT, 5pm PDT) for consideration in for current years award. Questions about the process may be sent to  

For the 2015 Savage award, up to two Awards of $750  will be made, one each in the Theory and Methods and Applied Methodology categories. Nominated theses have been evaluated by the Savage Award Committee. Four of the 8 finalists were invited to present their work at the 2016 ISBA World Meeting in Sardinia, with the winners announced at the meeting.

2015 Savage Award Committee

  • Ismael Castillo
  • Claudia Czado
  • Gauri Datta
  • Phil Dawid
  • Pierpaolo de Blasi
  • Pietro Dellaportas
  • Emily Fox
  • Tumiyasu Komaki
  • Long Nguyen
  • Omiros Papaspiliopoulos
  • Katya Scicciolo
  • Jim Scott 

Past Winners of the Savage Award


  • Theory & Methods
    Vinayak RaoMarkov chain Monte Carlo for continuous-time, discrete state systems.  University of Oxford; Yee Whye Teh, supervisor.
    Veronika Rockova (Honorable Mention), Bayesian variable selection in high-dimensional applications.  Erasmus Universiteit Rotterdam;  E. M. E. H. Lessafre and B. Lowenberg, supervisors
    Sayantan Banerjee (Honorable Mention), Bayesian inference for high dimensional models; convergence
    properties and computations issues
    .  North Carolina State University;  Subhashis Ghoshal, supervisor
  • Applied Methodology
    Christine PetersonBayesian graphical models for biological network inference. Rice University; Marina Vanucci
    Masanao Yajima (Honorable Mention), Bayesian Modeling of interactions of structured heterogeneous data.  University of California - Los Angeles; Jan de Leeuw and Donatello Telesca, supervisors


  • Theory & Methods
    Weining Shen, Adaptive Bayesian Function Estimation.  North Carolina State University; Subhasis Ghosal, supervisor.
    Depdeep Pati (Honorable Mention), Bayesian Nonparametric Modeling and Theory for Complex Data.  Duke University; David Dunsun, supervisor
  • Applied Methodology
    Ole Maneesoonthorn, Stochastic Volatility, Jumps and Variance Risk Premia: A Bayesian state space approach. Monash University;
    Osvaldo Anacleto (Honorable Mention), Bayesian Dynamic Graphical Models for High-Dimensional Flow Forescasting in Road Traffic Networks.  Open University; Catriona Queen and Paul Garthwaite, supervisors


  • Theory & Methods
    Anirban Bhattacharya, Bayesian Shrinkage in High Dimensions .  Duke University; David Dunson, supervisor.
    Bernardo Nipoti (Honorable Mention), Dependent Completely Random Measures and Statistical Applications .  University of Pavia; Antonio Lijoi, supervisor.
  • Applied Methodology
    Avishek Chakraborty, Modeling point patterns, measurement error and abundance for exploring species distributions . Duke University; Alan Gelfand, supervisor.
    Rebecca C. Steorts (Honorable Mention), Small Areas, Benchmarking, and Political Battles: Today's Novel Demands in Small-Area Estimation.  University of Florida; Malay Ghosh, supervisor. 


  • Theory & Methods
    Juan Carlos Martinez-Ovando, Contributions to Bayesian nonparametric modelling of time-series data.  University of Kent; Stephen G. Walker, supervisor.
     Gun Ho Jang (Honorable Mention), Invariant procedures in model checking for prior-data conflict and Bayesian analysis.  University of Toronto; Michael Evans, supervisor.
  • Applied Methodology
    Kaisey Mandel, Improving cosmological distances to illuminate dark energy: hierarchical Bayesian models for type Ia supernovae in the optical and near-infrared. Harvard University; Robert Kirshner, advisor.
    Fabian Scheipl (Honorable Mention), Spike-and-slab priors for function selection in structured additive regression models.  Ludwig-Maximilians-Universitat Munchen; Ludwig Fahrmeir, supervisor. 


  • Theory & Methods
    Julien CornebiseAdaptive Sequential Monte Carlo Methods. Université Pierre et Marie Curie - Paris 6, France; Eric Moulines, advisor.
    Daniel Williamson (Honorable Mention), Policy making using computer simulators for complex physical systems; Bayesian decision support for the development of adaptive strategies. Durham Univ, UK; Michael Goldstein, advisor.
  • Applied Methodology
    Ricardo LemosHierarchical Bayesian Methods for the Marine Sciences: Analyses of Climate Variability and Fish Abundance. University of Lisbon, Portugal; Henrique Cabral and Ramiro Neves, advisors.
    Robin Ryder (Honorable Mention), Phylogenetic Models of Language Diversification. University of Oxford, UK. Geoff Nicholls, advisor.


  • Theory & Methods
    James G. ScottBayesian Adjustment for Multiplicity. Duke Univ, USA; James Berger, advisor.
    Ryan Prescott Adams (Honorable Mention), Kernel Methods for Nonparametric Bayesian Inference of Probability Densities and Point Processes. University of Cambridge, UK; David MacKay, advisor.
  • Applied Methodology
    Emily B. FoxBayesian Nonparametric Learning of Complex Dynamical Phenomena. Massachusetts Institute of Technology, USA; Alan S. Willsky & John W. Fisher III, advisors.
    Matthew A. Taddy (Honorable Mention), Bayesian Nonparametric Analysis of Conditional Distributions and Inference for Poisson Point Processes. University of California at Santa Cruz, USA; Athanasios Kottas, advisor.


  • Theory & Methods
    Lorenzo TrippaSome extensions of the Polya urn scheme with Bayesian applications. L. Bocconi University (Milano, IT); Pietro Muliere, advisor.
  • Applied Methodology
    Alejandro JaraBayesian Semiparametric Methods for the Analysis of Complex Data. Katholieke Universiteit Leuven (BE); Emmanuel Lesaffre, Irene Gijbels and Geert Verbeke, advisors.
    Donatello Telesca (Honorable Mention), Bayesian Hierarchical Curve Registration. U Washington (Seattle, US); Lurdes Y.T. Inoue,advisor.
    Astrid Jullion (Finalist), Adaptive Bayesian P-splines models for fitting time-activity curves and estimating associated clinical measures in Positron Emission Tomography and Pharmacokinetic studies. Université Catholique de Louvain, Belgium; Philippe Lambert, advisor.


  • Theory & Methods
    Kostas KalogeropoulosBayesian Inference for Multidimensional Diffusion Processes. Athens Univ Econ and Business; Petros Dellaportas, advisor.
    Iain Murray (Honorable Mention), Advances in Markov chain Monte Carlo methods. Univ College London; Zoubin Ghahramani, advisor.
  • Applied Methodology
    Vladimir MininExploring Evolutionary Heterogeneity with Change-Point Models, Gaussian Markov Random Fields, and Markov Chain Induced Counting Processes. UCLA; Marc Suchard, advisor.
    Edoardo M. Airoldi (Honorable Mention), Bayesian Mixed-Membership Models of Complex and Evolving Networks. CMU; Stephen E. Fienberg and Kathleen Carley, advisors.


  • Theory & Methods
    Surya TokdarExploring Dirichlet Mixture and Logistic Gaussian Process Priors in Density Estimation, Regression and Sufficient Dimension Reduction. Purdue Univ; J.K. Ghosh, advisor.
    Pierpaolo de Blasi (Honorable Mention), Semiparametric models in Bayesian Event History Analysis. Bocconi U; Nils Lid Hjort & Pietro Muliere, advisors.
  • Applied Methodology
    Robert GramacyBayesian Treed Gaussian Process Models. UCSC; Herbie Lee, advisor.
    Carlos Carvalho (Honorable Mention), Structure and Sparsity in High-Dimensional Multivariate Analysis. Duke Univ; Mike West, advisor.


  • Theory & Methods
    Xinyi XuEstimation of High Dimensional Predictive Densities. Univ Penn; Ed George, advisor.
    Taeryon Choi (Honorable Mention), Posterior Consistency in Nonparametric Regression Problems under Gaussian Process Priors. CMU; Mark Schervish, advisor.
  • Applied Methodology
    Dimitris NicoloutsopoulosParametric and Bayesian Non-parametric Estimation of Copulas. UCL; Phil Dawid, advisor.
    Billy Amzal (Honorable Mention), Optimisation Bayésienne de Décisions et de Plans d'Expériences par Algorithmes Particulaires. Univ Paris Dauphine; Eric Parent & Christian Robert, advisors.


  • Theory and Methods
    Mario Trottini (Honorable Mention), Decision Models for Data Disclosure Limitation. CMU; Stephen Fienberg, advisor.
    Ramses Mena (Honorable Mention), Stationary Models using Latent Structures. Univ Bath; Stephen Walker, advisor.
  • Application Methodology
    Shane JensenStatistical Techniques for Examining Gene Regulation. Harvard Univ; Jun Liu, advisor.
    Jesus Palomo (Honorable Mention), Bayesian Methods in Bidding Processes. Rey Juan Carlos Univ; David Ríos Insua & Fabrizio Ruggeri, advisors.


  • Theory and Methods
    Chris PaciorekNon-stationary Gaussian Processes for Regression and Spatial Modeling. CMU; Mark Schervish, advisor.
  • Application Methodology
    Louis T. MarianoInformation Accumulation, Model Selection, and Rater Behavior in Constructed Response Student Assessments. CMU; Brian Junker, advisor.


  • Theory and Methods
    Nicolas ChopinApplications des Méthodes de Monte Carlo Séquentielles à la Statistique Bayésienne. Univ Paris Dauphine; Christian Robert, advisor.
  • Application Methodology
    Marc SuchardModel Building and Selection in Bayesian Phylogenetic Reconstruction. UCLA; Robert Weiss, advisor.


  • Theory and Methods
    Luis E. Nieto-BarajasBayesian Nonparametric Survival Analysis via Markov Processes. Univ Bath; Stephen Walker, advisor.
  • Application Methodology
    J. R. LockwoodEstimating Joint Distributions of Contaminants in U.S. Community Water System Sources. CMU; Mark Schervish, advisor.


  • Theory and Methods
    Peter Hoff (Co-winner), Constrained Nonparametric Estimation via Mixtures. Univ Wisconsin; Michael Newton, advisor.
    Tzee-Ming Huang (Co-winner), Convergence Rates for Posterior Distributions. CMU; Larry Wasserman, advisor.
  • Application Methodology
    Jeremy OakleyBayesian Uncertainty Analysis for Complex Computer Code. Univ Sheffield; Tony O'Hagan, advisor.
    Tim Hanson (Honorable Mention), Applied Bayesian Semiparametric Methods with Special Application to the Accelerated Failure Time Model and to Hierarchical Models for Screening UC Davis; Wes Johnson, advisor.


  • Theory and Methods
    Garrick L. WallstromConsistency and Strong Inconsistency of Inferences. Univ Minnesota; Joe Eaton, advisor.
  • Application Methodology
    Clare E. MarshallStatistical Methods for Institutional Comparisons. Cambridge Univ; David Spiegelhalter, advisor.
    Andrew S. Mugglin (Honorable Mention), Fully Model-Based Approaches for Spatially Misaligned Data. Univ Minnesota; Brad Carlin, advisor.


  • Antonietta MiraOrdering, splicing and splitting Monte Carlo Markov chains. Univ Minnesota; Luke Tierney, advisor.
  • Jaelong Lee (Honorable Mention) Semiparametric Bayesian analysis: selection models and meteorolgical applications. Purdue Univ; Jim Berger, advisor.


  • David DenisonSimulation Based Bayesian Non-parametric Regression Methods. Imperial College; Bani Mallick & Adrian Smith, advisors.
  • Juha Heikkinen (Honorable Mention), Bayesian Smoothing and Step Functions in Nonparametric Estimation of Curves and Surfaces. Univ Helsinki; Antti Penttinen & Elija Arjas, advisors.


  • Nariankadu D. ShyamalkumarContributions to Bayesian Nonparametrics and Bayesian Robustness. Purdue Univ; Jim Berger, advisor.
  • Eric Bradlow (Honorable Mention), A Hierarchical Latent Response Model for Ordinal Data with "No Answer" Responses. Harvard Univ; Alan Zaslavsky, advisor.
  • Max Chickering (Honorable Mention), Learning Bayesian Networks from Data. UCLA; David Heckerman, Richard Korf & Judea Pearl, advisors.
  • Andrea Piesse (Honorable Mention), Coherent Predictive Probabilities. Univ Canterbury; John Deely & Frank Lad, advisors.


  • Christopher K. Carter (Co-winner), On Markov Chain Monte Carlo for Linear State Space. Univ New South Wales; Robert Kohn, advisor.
  • Alyson Wilson (Co-winner) Statistical Models for Shapes and Deformations. Duke Univ; Valen Johnson, advisor.
  • Simon J. Godsill (Honorable Mention), The Restoration of Degraded Audio Signals.
  • Ming-Hui Chen (Honorable Mention), Monte Carlo Markov Chain Sampling for Bayesian Computation with Applications. Purdue Univ; Jim Berger & Bruce Schmeiseer, advisors.


  • Merlise A. ClydeBayesian Optimal Designs for Approximate Normality. Univ Minnesota; Kathryn Chaloner, advisor.
  • Marìa Del Carmen Fernandez-Llana (Honorable Mention), Estudios Sobre Robustez Bayesiana Global. Univ Autonoma de Madrid; Julian de la Horra, advisor.
  • Paul Gustafson (Honorable Mention), Local Sensitivity of Posterior Expectations. CMU; Larry Wasserman, advisor.
  • Debajyoti Sinha (Honorable Mention), Semiparametric Bayesian Analysis of Single and Multiple Time Event Data. Univ Rochester; W Jack Hall, advisor.


  • Charles Jeremy YorkBayesian Methods for the Analysis of Misclassified or Incomplete Multivariate Data.


  • Giovanni ParmigianiOptimal Scheduling of Inspections with an Application to Medical Screening Tests.


  • Valen E. JohnsonOn Statistical Image Reconstruction.


  • Michael David EscobarEstimating the Means of Several Normal Populations by Nonparametric Estimation of the distribution of the Means.


  • Peter E. RossiSpecification and Analysis of Econometric Production Models.


  • Mohan Delampady (Co-winner), Testing a Precise Hypothesis Interpreting P-Values from a Robust Bayesian Viewpoint.
  • S. Sivaganesan (Co-winner), Robust Bayesian Analysis with Contamminated Classes.
  • Herman K. van Dijk (Co-winner), Posterior Analysis of Econometric Models Using Monte Carlo Integration.


  • Peter Jamison LenkBayesian Nonparametric Predictive Distributions.


  • Luc Bauwens (Co-winner), Bayesian Full Information Analysis of Simultaneous Equation Models.
  • Peter C. Cranston (Co-winner), The Role of Time and Information in Bargaining.


  • Paul H. GarthwaiteAssessment of Prior Distributions for Normal Linear Models.


  • Soo Hong ChewTwo representation Theorems and Their Applications to Decision Theory.


  • Robert Kass (Co-winner), The Riemannian Structure of Model Spaces: A Geometrical Approach to Inference.
  • John P. O'Connor (Co-winner), A Certainty Equivlaent Based Metrization of Utility Function Space.


  • Paul MilgramThe Structure of Information in Competitive Bidding.


  • Kevin James McConwayThe Combination of Experts' Opinions in Probability Assessment: Some Theoretical Considerations.


  • Lorraine DeRobertisThe Use of Partial Prior Knowledge in Bayesian Inference.
  • José Bernardo (Honorable Mention), The Use of Information in the Design and Analysis of Scientific Experimentation.


  • Charles A. HoltBidding for Contracts.
  • Robert Shore (Honorable Mention), A Bayesian Approach to the Spectral Analysis of Stationary Time Series.