Fusion Learning & BFF

(Bayesian, frequentist and fiducial) Inferences and Statistical Foundations



Department of Statistics & DIMACS

Rutgers University


April 11-13, 2016

CoRE Auditorium, Busch Campus, Rutgers University




Monday, April 11

8:20am-9:00am: Registration/Breakfast

9:00am-9:15am: Welcome and Opening Remarks

                Regina Liu, Chair, Department of Statistics, Rutgers University

9:15am-10:55am: Chair, Regina Liu, Rutgers University

·         9:15am-9:55am: Min-ge Xie, Rutgers University

We R "BFF" (Best Friend Forever) on Road to BFF (Bayesian, Frequentist, Fiducial) Inferences

·         9:55am-10:25am:  Nils Hjort, University of Oslo

Confidence distributions for change points and regime shifts

·         10:25am-10:55am: Peter Song, University of Michigan

Confidence estimating functions

(10:55am-11:15am: Break)


11:15am-12:45pm: Chair, Min-ge Xie, Rutgers University

·         11:15am-11:45am: Glenn Shafer, Rutgers University

What does “frequentist” mean?

·         11:45am-12:45pm (Keynote): Brad Efron, Stanford University

Confidence densities, uninformative priors, and the bootstrap

·         Discussant: Cun-Hui Zhang, Rutgers University

 (12:45pm-2:00pm: Lunch and Poster Session)


2:00pm-3:40pm: Chair, Rong Chen, Rutgers University

·         2:00pm-2:10pm: DIMACS Director’s Welcome, Rebecca Wright, Director of DIMACS

·         2:10pm-2:40pm: Andrew Gelman, Columbia University

Taking Bayesian inference seriously

·         2:40pm-3:10pm: Hari Iyer and Steven Lund, National Institute of Standards and Technology (NIST)

A problem in forensic science? Whose prior, whose Bayes factor, and who are you kidding?

·         3:10pm-3:40pm:  David Draper, UC-Santa Cruz

Rigorizing and extending the Cox–Jaynes derivation of probability: implications for statistical practice

(3:40pm-3:55pm: Break)


3:55pm-5:30pm: Chair, William Strawderman, Rutgers University

·         3:55pm-4:55pm (Keynote): Jim Berger, Duke University

The use of rejection odds and rejection ratios in testing hypotheses

·         Discussant: Dongchu Sun, University of Missouri

5:00pm-5:30pm: Workshop Mixer/Poster Session


Tuesday, April 12

8:20am-8:45am: Registration/Breakfast

8:45am-10:30am: Chair, Dennis Cox, Rice University

·         8:45am-9:30am: Xiao-Li Meng, Harvard University

Let’s believe belief functions: a paradigm for multi-resolution probabilistic inference

·         9:30am-10:00am:  Ryan Martin, University of Illinois-Chicago

On beliefs, validity, and the foundations of statistics

·         10:00am-10:30am: Jan Hannig, UNC-Chapel Hill

New challenges in generalized fiducial inference

(10:30am-10:50am: Break)


10:50am-12:20pm: Chair, John Kolassa, Rutgers University

·         10:50am-11:20am: Nozer Singpurwalla, The City University of Hong Kong

On the missing F in BFF

·         11:20am-11:50am: Ulrich Müller, Princeton University

Credibility of confidence sets in nonstandard econometric problems

·         11:50am-12:20pm: Ying Hung, Rutgers University

A sequential split-conquer-combine approach for analysis of big spatial data using confidence distributions

(12:20pm-1:40pm: Lunch and Poster Session)


1:40pm-3:25pm: Chair, Dan Yang, Rutgers University

·         1:40pm-2:25pm: Don Fraser, University of Toronto

What can we expect from distributions for parameters

·         2:25pm-2:55pm: Hongzhe Li, University of Pennsylvania

Sparse Simultaneous Signal Detection and Its Applications in Genomics

·         2:55pm-3:25pm: Mounir Mesbah, Université Pierre et Marie Curie, Paris 6

The backward reliability curve and its practical usefulness

(3:25pm-3:40pm: Break)


3:40pm-5:20pm: Chair, Han Xiao, Rutgers University

·         3:40pm-4:05pm: Michael Fay, National Institute of Allergy and Infectious Diseases

Combining one-sample confidence interval procedures for valid non-asymptotic inference in the two-sample case

·         4:05pm-4:30pm: Dungang Liu

Fusion Learning: combining of inferences from diverse sources using data depth and confidence distribution

·         4:30pm-4:55pm: Keli Liu, Stanford University

Can big data help us better understand statistical foundations?

·         4:55pm-5:20pm: Harry Crane, Rutgers University

Edge exchangeability: a new foundation for modeling network data

5:25pm-6:00pm: Poster session


Wednesday, April 13

8:20am-8:50am: Registration/Breakfast

8:50am-10:35am: Chair, Lee Dicker, Rutgers University

·         8:50am-9:35am: Ed George, University of Pennsylvania

Fast Bayesian Factor Analysis via Automatic Rotations to Sparsity

·         9:35am-10:05am: Dipak Dey, University of Connecticut

Bayesian inference using Bregman divergence measures

·         10:05am-10:35am: Veronika Rockova, University of Pennsylvania

The Spike-and-Slab LASSO

 (10:35am-10:55am: Break)


10:55am-12:25pm: Chair, Xiao-Li Meng, Harvard University

·         10:55am-11:25am: Sam Weeranhandi, Pfizer

Still researching on asymptotic methods? Try generalized inference!

·         11:25am-12:25pm (Featured video talk): Sir David R. Cox, Oxford University

Data-based distributions for unknown parameters: always, sometimes, never?

·         Discussant: Nancy Reid, University of Toronto

 (12:25pm-1:30pm: Lunch)


1:30pm-3:10pm: Chair, Regina Liu, Rutgers University

·         1:30pm-1:55pm: Arne Bathke, University of Salzburg

Synthesizing information and making local conclusions: multivariate inference, multiple tests, and not so many assumptions

·         1:55pm-2:20pm: Ming-Yen Cheng, National Taiwan University

A new test for functional one-way ANOVA with application to ischemic heart screening

·         2:20pm-2:45pm: Paul Edlefsen, Fred Hutchinson Cancer Research Center

The general univariate Dempster-Shafer model and its survival analysis counterpart for evaluating HIV-1 vaccine efficacy when censorship is not random

·        2:45pm-3:10pm: Benjamin Holcblat and Steffen Grønneberg, BI Norwegian Business School

Statistical inference theories, multiple uses of the same data, and past-realized data

3:10pm-3:20pm: Closing Remarks/Discussions



Organizing Committee:

Harry Crane, Lee Dicker, Ying Hung, Regina Liu (co-chair), John Kolassa, William Strawderman, Han Xiao, Minge Xie (co-chair), Dan Yang


For more information: http://www.stat.rutgers.edu