Please Note: This seminar will be given online.
University of Michigan
Link to join seminar: Hosted on Zoom
Random Effect Models in High Dimensions
Most of the literature on high-dimensional data focuses on independent and identically distributed data. While theoretically elegant and tractable, this limits the applicability of classical tools to many practical settings encountered in the sciences; for example, in longitudinal studies, individuals are observed over time, creating dependence between the measurements. Thus, to account for dependence, heterogeneity needs to be incorporated in both the modeling and the analysis. In this talk, I will discuss some of our recent work on estimation and inference in two high-dimensional models with random effects: (i) linear mixed effects models and (ii) varying coefficient models with functional random effects.