Bayesian, Fiducial, and Frequentist (BFF) Conferences


This workshop was held at Penn Pavilion on the campus of Duke University.


The Bayesian, Fiducial, and Frequentist (BFF) conferences began in 2014 as a means of facilitating scientific exchange among statisticians developing new BFF methodologies. These conferences offer an opportunity to examine different statistical paradigms, and compare various methodologies. This year BFF6 was held jointly with the 2018-2019 Model Uncertainty: Mathematical and Statistical (MUMS) program at the Statistical and Applied Mathematical Sciences Institute (SAMSI), and in particular MUMS working groups on the Foundations of Model Uncertainty and on Data Fusion.

** Notice of Consent **

SAMSI values the proprietary and intellectual property of our participants. The materials presented at our various workshops and programs are in high demand by event participants and the applied mathematics and statistics community that comprise our audience. Therefore, we encourage all of our invited speakers to share their materials, as appropriate, in order to pass along the valuable research that is being done in your field of study and is a focus of this event. In addition, unless SAMSI is give written approval from our speakers we ARE NOT authorized to share the materials presented at this event.

Please click HERE to complete a SAMSI Consent form for this event. SAMSI appreciates your time and willingness to share this valuable content with others and we hope you enjoy this event!
For any questions or concerns about our consent policy, please contact us at:

Keynote Lecture

James Scott, Associate Professor of Statistics, Dept. of Information, Risk, and Operations Management, McCombs School of Business, University of Texas at Austin

Date/Location: April 29, 2019, 7-8pm at the Penn Pavilion on the campus of Duke University

Title: How People and Machines are Smarter Together

Description:Our world today is filled with complex machines and technologies that have the capability to learn and adapt. From smart phones to driverless cars, these machines understand how humans interact with the world by using the wealth of data made available to them through our interactions every day. This technology is helping to reshape our future the same way the Industrial Revolution remade the 19th century.

In the book AIQ: How People and Machines are Smarter Together, James Scott and Nick Polson address how mathematics drives the language spoken by the artificially intelligent machines in our world today. The book takes us through history and teaches us about data, probability and better thinking. In addition, the book shows how ideas from the past are being used to help intelligent machines to adapt in the modern age using big data. Understanding these technologies can help us to overcome some of our cognitive shortcomings and help us to lead happier and more productive lives.

In this lecture, James Scott talked about his new book and discusses the concepts addressed in it.

** A short course will be presented on April 28; research contributions will be presented April 29-May 1. **.

Schedule and Supporting Media

Printed Schedule
Titles and Abstracts
Poster Titles
Keynote Lecture Flyer
Participant List

Sunday, April 28, 2019
Penn Pavilion, Duke University, Durham, NC

Description Speaker Slides
Short Course on Bayesian Model Uncertainty: Model Uncertainty: A Review Anabel Forte, University of Valencia
Short Course on Data Fusion
Fusion Learning and BFF Approaches Regine Liu, Rutgers University
Min-ge Xie, Rutgers University

Monday, April 29, 2019
Penn Pavilion, Duke University, Durham, NC

Description Speaker Slides
Welcome and Introductory Information Cynthia Rudin, Duke University
Keynote Lecture: Inference Meets Computation: Dynamical, Stochastic and Economic Perspectives Michael Jordan, University of California, Berkeley
Invited Session: BFF Interfaces
Maximum Entropy Derived and Generalized Under Idempotent Probability to Address Bayes-Frequentist Uncertainty and Model Revision Uncertainty David Bickel, University of Ottawa
Multidimensional Monotonicity Discovery with MBART Ed George, University of Pennsylvania
Calibration of Probability Forecasts Vladimir Vovk, Royal Holloway, University of London
Keynote Lecture: Statistical Sparsity Peter McCullagh, University of Chicago
Invited Session: Uncertainty Quantification for Bayesian Nonparametrics
Multiscale Analysis of BART Veronika Rockova, University of Chicago
Uncertainty Quantification for Bayesian Survival Analysis Stéphanie van der Pas, Leiden University
Coverage of Credible Intervals for Monotone Regression Subhashis Ghosal, N.C. State University
Panel Session Moderator: Nancy Reid, University of Toronto Statistics
Panelists Glenn Shafer, Rutgers Statistics/Business School
Peter Song, Michigan Biostatistics
Naveen Narisetty, UIUC Statistics
Ruobin Gong, Rutgers Statistics
Poster Session and Reception
Public Lecture: Discussing his new book “AIQ: How People and Machines Are Smarter Together” James Scott, University of Texas, Austin

Tuesday, April 30, 2019
Penn Pavilion, Duke University, Durham, NC

Description Speaker Slides
Keynote Lecture: Can a Fiducial Phoenix Rise from the Ashes? Phil Dawid, Cambridge University
Invited Session: SAMSI Developments on Data Fusion
Generalized Probabilistic Principal Component Analysis of Correlated Data Mengyang Gu, Johns Hopkins University
Are Reported Likelihood Ratios Well Calibrated? Jan Hannig, University of North Carolina, Chapel Hill
Bayesian Analysis for Misaligned Regions and Applications in Cancer Mortality Dongchu Sun, University of Missouri
Keynote Lecture: Spatially Informed Variable Selection Priors and Applications to Large-scale Data Marina Vannucci, Rice University
Invited Session: SAMSI Developments in Model Uncertainty
Variable Selection in the Discrepancy Function Associated with a Simulator Pierre Barbillon, SAMSI and UMR MIA-Paris, AgroParisTech, INRA
Including Factors in Bayesian Variable Selection Problems Gonzalo Garcia-Donato, Universidad de Castilla-La Mancha
Model Selection in the Context of Computer Models Rui Paulo, ISEG, Technical University of Lisbon
Student Invited Session
Objective Bayesian Analysis for a 2 x 2 Contingency Table John Snyder, University of Missouri
Inference on Treatment Effects after Model Selection Jingshen Wang, University of Michigan
The EAS Approach for Graphical Selection Consistency in Vector Autoregression Models Jonathan Williams, University of North Carolina, Chapel Hill

Wednesday, May 1, 2019
Penn Pavilion, Duke University, Durham, NC

Description Speaker Slides
Keynote Lecture: Selecting Important Features in Presence of Correlation—a story from Genetics Chiara Sabatti, Stanford University
Invited Session: Philosophical Perspective on Model Uncertainty
A New, Truth-directed Explanation of Ockham’s Razor in Model Inference Kevin Kelly, Carnegie Mellon University
Unifying Ockham’s Razor Leah Henderson, University of Groningen
Models as Tools not Mirrors: Crossover Themes from the Philosophy of Science Alisa Bokulich, Boston University
Final Discussion
Closing Remarks Ruobin Gong, Rutgers University
Jan Hannig, University of North Carolina, Chapel Hill

Questions: email