A Z-estimation system - a LEGO approach for systematically developing models
This presentation will feature Dr. Kate Hu, Assistant Professor in the Department of Statistics at Ohio State.
About
From the speaker:
In this talk, I will introduce a Z-estimation system—a modular, “Lego-like” approach for systematically developing models and conducting asymptotic analysis. I will explain the core principles behind this system, how it differs from the classical Taylor expansion method, and what makes it inherently systematic. Using this framework, I will demonstrate how to derive asymptotic variances for estimators in general parametric and semiparametric models across various types of data, including random sampling, two-phase sampling, and two-phase sampling with auxiliary information. To make the ideas concrete, I will apply the results to additive hazards models. Finally, I will discuss current limitations, future directions for software and model development, and potential student projects.
The Biostatistics seminar series invites researchers from across the nation to discuss methodological research and its implications for a variety of health issues.
Contact
Andy Ni