Bohrer Lecture: Judy Wang, George Washington University

Judy Wang
Huixia Judy Wang
George Washington University

BIO: Huixia Judy Wang is a statistician who works as a professor of statistics at George Washington University. Topics in her research include quantile regression and the application of biostatistics to cancer.  Wang graduated from Fudan University in 1999 and earned a master's degree from Fudan in 2002. She completed her Ph.D. in statistics in 2006 from the University of Illinois at Urbana–Champaign. Her dissertation, Inference on Quantile Regression for Mixed Models with Applications to GeneChip Data, was supervised by Xuming He. She joined the statistics faculty at North Carolina State University in 2006 and moved to George Washington University in 2014. From 2018, she has been serving as program director for the Statistics Program at the National Science Foundation.

Talk Title: Flexible Probabilistic Prediction for Non-Gaussian Spatial Processes

Talk Abstract: Many existing methods for analyzing spatial data rely on the Gaussian assumption, which is violated in many applications such as wind speed, precipitation and COVID mortality data. In this talk, I will discuss several semiparametric approaches for probabilistic prediction of non-Gaussian spatial data. First, I will present a copula-based multiple indicator kriging (CMIK) model for the analysis of non-Gaussian spatial data. The algorithm is computationally simple since it models the marginal distribution and the spatial dependence separately. Next, I will introduce a deep multiple indicator kriging (DMIK) method for univariate and bivariate spatial processes. The method is based on thresholding the spatial observations at a given set of quantile values within a deep neural network framework. The DMIK method overcomes the computational challenge in the CMIK for analyzing large datasets. The developed methods do not require any parametric assumptions on the marginal distribution and are thus more flexible than Gaussian-based methods. These methods will also provide convenient ways to construct both point and interval predictions based on the estimated conditional quantiles. Finally, I will briefly discuss semiparametric approaches for analyzing spatio-temporal and count spatial data, and present some numerical results.

 

Wijsman Lecture: Bodhisattva Sen - Columbia University

Bodhisattva Sen
Bodhisattva Sen
Wikipedia

BIO: I am a Professor of Statistics at Columbia University, New York.  My core statistical research centers around nonparametrics and large sample theory -- nonparametric function estimation (with special emphasis on shape constrained estimation), likelihood and bootstrap based inference in (non-standard) parametric and nonparametric models, optimal transportation and applications in Statistics, etc. I am also actively involved in interdisciplinary research, especially in astronomy.

I completed my Ph.D in Statistics from University of Michigan, Ann Arbor, in 2008.  Prior to this I was a student at the Indian Statistical Institute, Kolkata, where I received Bachelors (B.Stat.) and Masters (M.Stat) in Statistics.

Apart from Statistics, I like to travel.  I am also interested in photography and cricket.

Talk Title: Extending the Scope of Nonparametric Empirical Bayes
Talk Abstract: In this talk we will describe two applications of empirical Bayes (EB) methodology. EB procedures estimate the prior probability distribution in a latent variable model or Bayesian model from the data. In the first part we study the (Gaussian) signal plus noise model with multivariate, heteroscedastic errors. This model arises in many large-scale denoising problems (e.g., in astronomy). We consider the nonparametric maximum likelihood estimator (NPMLE) in this setting. We study the characterization, uniqueness, and computation of the NPMLE which estimates the unknown (arbitrary) prior by solving an infinite-dimensional convex optimization problem. The EB posterior means based on the NPMLE have low regret, meaning they closely target the oracle posterior means one would compute with the true prior in hand. We demonstrate the adaptive and near-optimal properties of the NPMLE for density estimation, denoising and deconvolution.

In the second half of the talk, we consider the problem of Bayesian high dimensional regression where the regression coefficients are drawn i.i.d. from an unknown prior. To estimate this prior distribution, we propose and study a "variational empirical Bayes" approach — it combines EB inference with a variational approximation (VA). The idea is to approximate the intractable marginal log-likelihood of the response vector --- also known as the "evidence" --- by the evidence lower bound (ELBO) obtained from a naive mean field (NMF) approximation. We then maximize this lower bound over a suitable class of prior distributions in a computationally feasible way. We show that the marginal log-likelihood function can be (uniformly) approximated by its mean field counterpart. More importantly, under suitable conditions, we establish that this strategy leads to consistent approximation of the true posterior and provides asymptotically valid posterior inference for the regression coefficients.

 

Past Bohrer Workshop Keynote Speakers

  • 1994 Mark Schervish, Carnegie Mellon University
  • 1995 Dan Naiman, Johns Hopkins University
  • 1997 Ross Leadbetter, University of North Carolina
  • 1998 Dennis Karney, University of Kansas
  • 1999 Erich Lehmann, University of California Berkeley
  • 2000 David Bartholomew, London School of Economics
  • 2001 Gary Koch, University of North Carolina
  • 2002 Robert Serfling, University of Texas Arlington
  • 2003 Peter Bickel, University of California Berkeley
  • 2004 Peter Imrey, Cleveland Clinic Foundation
  • 2005 John Marden, University of Illinois
  • 2006 Raymond Carroll, Texas A&M University
  • 2007 Mary Ellen Bock, Purdue University
  • 2008 Ker-Chau Li, UCLA
  • 2010 Zhiliang Ying, Columbia University
  • 2011 Minge Xie, Rutgers University
  • 2012 Xuming He, University of Michigan
  • 2013 Yuhong Yang, University of Minnesota
  • Sky Andrecheck, Cleveland Indians
  • 2014 Hua-Hua Chang, University of Illinois at Urbana-Champaign
  • 2015 Cun-Hui Zhang, Rutgers University
  • 2016 Lawrence Brown, University of Pennsylvania
  • 2017 Edward George, University of Pennsylvania
  • 2018 Regina Liu, Rutgers University
  • 2019 - 2020 Cancelled
  • 2021 Liza Levina, University of Michigan
  • 2022 Vijay Nair, University of Michigan
  • 2023 Andrew R. Barron, Yale University

Past Wijsman Lecturers

  • 2014 Zhiliang Ying, Columbia University
  • 2015 John Lafferty, University of Chicago
  • 2016 Xihong Lin, Harvard University
  • 2017 Nancy Reid, University of Toronto
  • 2018 Michael Kosorok, University of North Carolina at Chapel Hill
  • 2019 - 2020 Cancelled
  • 2021 Dean Foster, Amazon
  • 2022 Dawn Woodard, Uber
  • 2023 Jelena Bradic, University of California San Diego