Title：Optimal Subsampling for High-Dimensional Partially Linear Models via Machine Learning Methods

Abstract：In this paper, we explore optimal subsampling strategies for estimating the parametric regression coefficients in partially linear models with unknown nuisance functions involving high-dimensional and potentially endogenous covariates. To address model misspecifications and the curse of dimensionality, we leverage flexible machine learning (ML) techniques to estimate the unknown nuisance functions.By constructing an unbiased subsampling Neyman-orthogonal score function, we eliminate regularization bias. A two-step algorithm is then used to obtain appropriate ML estimators of the nuisance functions, mitigating the risk of overfitting. Using martingale techniques, we establish the unconditional consistency and asymptotic normality of the subsample estimators. Furthermore, we derive optimal subsampling probabilities, including A-optimal and L-optimal probabilities as special cases.The proposed optimal subsampling approach is extended to partially linear instrumental variable models to account for potential endogeneity through instrumental variables. Simulation studies and an empirical analysis of the Physicochemical Properties of Protein Tertiary Structure dataset demonstrate the superior performance of our subsample estimators.

Speaker：Lei Wang is a Professor and PhD supervisor at the School of Statistics and Data Science, Nankai University, and a member of the “One Hundred Young Academic Leaders” program. His research interests include statistical learning and complex data analysis. He has published numerous papers in leading statistical journals such as Biometrika, Journal of Machine Learning Research (JMLR), IEEE Transactions on Information Theory, Annals of Applied Statistics (AOAS), Bernoulli, Journal of Computational and Graphical Statistics (JCGS), and Statistica Sinica. He has presided over three projects funded by the National Natural Science Foundation of China and one project funded by the Tianjin Natural Science Foundation.

Date：9:30–10:30 a.m., Friday, November 21, 2025

Venue: Lecture Hall 208, School of Mathematics, Shouxi Lake Campus, Yangzhou University

Organizer：School of Mathematics

Inviter：Zhensheng Huang, Jun Jin

Students and teachers are welcome.