SDS Seminar Series – Richard Samworth, University of Cambridge
Apr
18
2025
Apr
18
2025
Description
The Spring 2025 SDS Seminar Series continues on April 18th from 2:00 p.m. to 3:00 p.m. with Dr. Richard Samworth (Professor, Department of Pure Mathematics and Mathematical Statistics and Director, Statistical Laboratory, University of Cambridge). This event is in-person in the Avaya Room (POB 2.302).
Title: How Should We Do Linear Regression?
Abstract: In the context of linear regression, we construct a data-driven convex loss function with respect to which empirical risk minimisation yields optimal asymptotic variance in the downstream estimation of the regression coefficients. Our semiparametric approach targets the best decreasing approximation of the derivative of the log-density of the noise distribution. At the population level, this fitting process is a nonparametric extension of score matching, corresponding to a log-concave projection of the noise distribution with respect to the Fisher divergence. The procedure is computationally efficient, and we prove that our procedure attains the minimal asymptotic covariance among all convex M-estimators. As an example of a non-log-concave setting, for Cauchy errors, the optimal convex loss function is Huber-like, and our procedure yields an asymptotic efficiency greater than 0.87 relative to the oracle maximum likelihood estimator of the regression coefficients that uses knowledge of this error distribution; in this sense, we obtain robustness without sacrificing much efficiency. Numerical experiments confirm the practical merits of our proposal.
Other Events in This Series
Oct
4
2024
SDS Seminar Series – Huiyan Sang, Texas A&M University
GS-BART: Graph Split Additive Decision Trees for Spatial and Network Data
2:00 pm – 3:00 pm • In Person
Speaker(s): Huiyan Sang
Oct
11
2024
SDS Seminar Series – Mingyuan Zhou, University of Texas at Austin
Building Faster, Better, and Safer Deep Generative Models via Score Identity Distillation
2:00 pm – 3:00 pm • In Person
Speaker(s): Mingyuan Zhou
Oct
18
2024
SDS Seminar Series – Sherry Zhang, University of Texas at Austin
Pivoting between Space and Time: Spatio-Temporal Analysis with Cubble
2:00 pm – 3:00 pm • In Person
Speaker(s): Sherry Zhang
Oct
25
2024
SDS Seminar Series – Matt Koslovsky, Colorado State University
Sparse Dirichlet-Multinomial Models
2:00 pm – 3:00 pm • In Person
Speaker(s): Matt Koslovsky
Nov
1
2024
SDS Seminar Series – Aaditya Ramdas, Carnegie Mellon University
A Game-Theoretic Theory of Statistical Evidence
2:00 pm – 3:00 pm • In Person
Speaker(s): Aaditya Ramdas
Nov
8
2024
SDS Seminar Series – Myungsoo Yoo, University of Texas at Austin
Dynamic Spatio-Temporal Model Integrating Physics for Fire Front Propagation
2:00 pm – 3:00 pm • In Person
Speaker(s): Myungsoo Yoo
Nov
15
2024
SDS Seminar Series – Rafael Irizarry, Harvard University
Twenty-Five Years of Data Science: Music, Genomics, and Public Health Surveillance
2:00 pm – 3:00 pm • In Person
Speaker(s): Rafael Irizarry
Mar
7
2025
SDS Seminar Series - Arun Kuchibhotla, Carnegie Mellon University
Adaptive Inference Techniques for Some Irregular Problems
2:00 pm – 3:00 pm • In Person
Speaker(s): Arun Kuchibhotla
Mar
28
2025
SDS Seminar Series – Po-Ling Loh, University of Cambridge
Differentially Private M-estimation via Noisy Optimization
2:00 pm – 3:00 pm • In Person
Speaker(s): Po-Ling Loh
