Bertrand Clarke

    Bertrand Clarke
    Professor of Statistics

    Department of Medicine, Department of Epidemiology and Public Health in the Faculty of Medicine. Also, last but not least, Center for Computational Science.
    Miller School of Medicine
    University of Miami
    1120 NW 14th Street, CRB 1055 (C-213)
    Miami, FL 33136
    e-mail: bclarke2@med.miami.edu

    Degrees and position

    • B.Sc. (1984), University of Toronto
    • Ph.D. (1989), University of Illinois
    • Assistant Professor, Purdue University (1989-92)
    • The University of British Columbia (1992-2009)

    I was Program Chair for Risk Analysis at JSM 2009. I was regional representative for the SSC for 2006/7 and AE for JASA (2005/7).
    Currently I’m AE for JSPI and for Statistical Papers.

    Research Interests

    I first cut my teeth on research in Mathematical Biology with Jay Mittenthal. It sealed my fate: The research world grabbed me by the scruff of the neck and just never let go. At the time I thought I would be a mathematician. Under the influence of my thesis advisor Andrew Barron, and my major undergrad influence, Don Fraser, I drifted into Statistics. Drifted may not be the correct word – but I did like mathematical biology, and I did like information theory, and I fell in love with predictive ideas in general while studying Fraser’s Structural Statistics on an NSERC summer undergrad award in 1984. So, Statistics was really the only place I could go to satisfy my intellectual cravings.

    I have some early heroes: Don Fraser, Andrew Barron, Jay Mittenthal, Jayanta Ghosh…and a few others. More about them later.

    So what happened? I got my PhD in a mix of information theory and Bayesian asymptotics. That hooked me on central limit theory (very mysterious…still astonishes me when I think about what it means), prior selection (reference priors mostly, but any kind of objective prior obtained from optimization), and thinking of data as information in the sense of strings of zeros and ones a la Shannon theory. All this went on well and good – I was exploring various fields and doing the research that seemed fun.

    Then somehow my earlier interest in prediction came back in a big way. Partially it was a new grad student who was interested in it. Partially it was that I went on sabbatical to UCL 1997-1998 and spent an academic year working with Phil Dawid, one of the most original thinkers the statistical world has. Mostly, it was just that none of the branches of statistics I’d seen were fully satisfying since they neglected so many sources of uncertainty and seemed not to be as closely tied to real data as I thought they should be.
    I still love research of all sorts…I’ve always been very distractable because the whole damn world is so interesting.

    Here’s what you find on the rest of my webpage:

    My CV

    My work on Theory and Methods
    Here’s my stuff on prior selection, central limit theory, model selection, robustness amongst other topics.

    My work on Mathematical Biology
    This is a little dated, but we focused on network modeling and even develop some software.

    My work on Prequential and Predictive statistics
    This is my current fascination especially from a Data Mining and Machine Learning perspective.

    Other Publications
    This is a whole bunch of other publications that don’t really fit anywhere. You’ll find the cover of my book (yayyy!) and an errata sheet (*sigh*). There is a copy of a Festschrift for J. K. Ghosh and a few unpublished manuscripts. There’s even some nontechnical material.

    Software

    This has the complete material for Netscan a suite of material for generating networks to satisfy constraints. There’s a lot of work here and it seems suddenly relevant again.