On novel application of classical linear regression in biomedical study

Abstract

In this talk, I will present two methodological developments grounded in classical linear regression frameworks. The first concerns inference on the association between exposure mixtures and continuous health outcomes, and the second addresses modeling strategies for resting energy expenditure.  For both problems, we formulate estimation procedures under working-model assumptions and develop corresponding hypothesis tests. Theoretical properties of the proposed methods are established, and numerical studies are provided to evaluate finite-sample performance and illustrate practical implementation.

Please click here for participating the talk online.