2026
Student Seminar: Junhan Li
2:30–3:00 pm Jones 111
Wednesday, November 11, 2026, at 2:30 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Junhan Li, Department of Statistics, The University of Chicago
“Uniform Confidence Bands for the Empirical Pricing Kernel”
Student Seminar: Zi'ang Yuan
2:00–2:30 pm Jones 111
Wednesday, November 11, 2026, at 2:00 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Zi’ang Yuan, Department of Statistics, The University of Chicago
“Rerandomization under Interference”
Student Seminar: Vedant Pathak
1:30–2:00 pm Jones 111
Wednesday, November 11, 2026, at 1:30 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Vedant Pathak, Department of Statistics, The University of Chicago
“The SphericalTNBBeta Distribution and its Applications to Variational Autoencoders”
Student Seminar: Xin Xie
10:30–11:00 am Jones 111
Wednesday, November 11, 2026, at 10:30 AM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Xin Xie, Department of Statistics, The University of Chicago
“Controlling Tail-Risk Amplification under Market Stress (A Bounded Dynamic Model for VaR and Expected Shortfall)”
Statistics Colloquium: Bin Yu
11:30 am–12:30 pm Jones 303
Bin Yu
Department of Statistics, Electrical Engineering and Computer Sciences, and Center for Computational Biology
University of California, Berkeley
Title: Not All Interpretations Are Created Equal: Principled Evaluation of AI Interpretability
Abstract: As AI transitions to agentic sociotechnical deployments, uninterpretable models pose serious threats to reliability, public trust, and contestability. However, not all interpretation methods are created equal: for interpretability to enhance deployment safety and effectiveness, interpretation methods must be evaluated as a whole, rather than validated through isolated interpretation instances.
In this paper, we advocate elevating the evaluation of interpretation methods as a core assessment metric for AI system deployment, with clearly stated principles. To this end, we propose TRUST, a principled framework that builds upon Veridical Data Science (VDS) and the Predictive, Descriptive and Relevance (PDR) foundations. To differentiate interpretation methods effectively, TRUST systematically grades interpretation methods across five progressive tiers: Task-Aligned PDR Foundations, Rigorous Epistemic Grounding, Unifying Semantic Understanding, Scope Transferability, and Targeted Prospective Validation.
Student Seminar: Daniel Purizhansky
10:00–10:30 am Jones 111
Monday, November 9, 2026, at 10:00 AM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Daniel Purizhansky, Department of Statistics, The University of Chicago
“Generative Models of Heterogeneity in Compositional Microbiome Data and the Batch-Corrected Latent Space”
Student Seminar: Ruochen Wang
2:00–2:30 pm Jones 111
Wednesday, November 4, 2026, at 2:00 PM, in Jones 111, 5747 S. Ellis Avenue
Master’s Thesis Presentation
Ruochen Wang, Department of Statistics, The University of Chicago
“Training Window Length and Partial Pooling for Prescription Forecasting: A Longitudinal Count Study of Empagliflozin Prescribing in England, 2018–2026”
Statistics Colloquium: Michael Sobel
11:30 am–12:30 pm Jones 303
Michael Sobel
Department of Statistics
Columbia University
Title: Measuring Public Institutions and their Agents: Disparities and Agreement, with Application to Asylum Decisions in the U.S. Immigration Courts
Abstract: Public institutions derive their legitimacy from the professionalism and impartiality of their agents. In order to maintain standards of conduct, the performance of an agent is often compared to the performance of her peers. But public institutions typically assign agents to different tasks, and as a result, naive comparisons may reflect the difficulty of the assignment and not an agent’s. To address this concern, we develop measures of disparity and agreement that compare an agent’s performance with the potential performance of randomly selected peers who could have been assigned the same set of tasks. We then give realistic conditions under which disparity is identified from the marginal distribution of each agent’s actions. We also show how agreement, which is generally unidentified when agents are assigned to different tasks, can be bounded or estimated under sensitivity assumptions about the dependence between potential outcomes. Finally, we use a nonparametric Bayesian model to examine applications for asylum filed between 2007 and 2014 in the widely studied and often maligned U .S. immigration courts. Our results do not support the picture of widespread partiality asserted in previous empirical work. At both the judge and court level, we find limited disparity and substantial potential agreement.
Students Seminar: Yating Liu
10:00 am–12:00 pm Jones 111
Tuesday, October 27, 2026, at 10:00 AM, in Jones 111, 5747 S. Ellis Avenue
Dissertation Defense Presentation
Yating Liu, Department of Statistics, The University of Chicago
“Learning from Text to Biological Data: The Trustworthy Inference using Robust Latent Structure”
Statistics Colloquium: Jing Lei
11:30 am–12:30 pm Jones 303
Jing Lei
Department of Statistics & Data Science
Carnegie Mellon University
Title: Evaluating Black-Box Classifiers via Stable Adaptive Two-Sample Inference
Abstract: Black-box machine learning methods predict well but are hard to validate. This talk develops adaptive, sample-efficient inference that uses flexible machine learning tools to test black-box predictions, with conditional distribution testing as the unifying problem. First, a weighted conformal rank-sum statistic turns aggregated conformal p-values into a test whose target, the Neyman-Pearson-Gini divergence, connects rank-sum statistics, ROC curves, and total variation. Second, two-sample U-statistic orthogonalization and cross-fitting remove the bias from estimating density-ratio weights, so the data can be reused without sample splitting. Third, for validating a fitted classifier, a cross-fitted distinguisher with a “safe-side bias” yields valid inference even though the statistic targets a data-dependent lower bound of the divergence. Applications show that classifiers with similar accuracy can differ substantially in goodness of fit.