QSU Research Methods Seminar: Robert Tibshirani, Ph.D

Tuesday, April 1, at 12:00 pm PST

Join the seminar at 3180 Porter Dr. (combined Room B107 & B136)

*Lunch will be provided*

Zoom Link:
Meeting ID:  996 4643 5133
Password: 593858

Robert Tibshirani, Ph.D.
Professor of Biomedical Data Science and of Statistics
Stanford University

Univariate-guided Sparse Regression

In this talk Dr. Tibshirani will introduce “UniLasso”, a novel statistical method for sparse regression. This two-stage approach preserves the signs of the univariate coefficients and leverages their magnitude. Both of these properties are attractive for stability and interpretation of the model. Through comprehensive simulations and applications to real-world datasets, we demonstrate that UniLasso outperforms Lasso in various settings, particularly in terms of sparsity and model interpretability. We prove asymptotic support recovery and mean-squared error consistency under a set of conditions different from the well-known irrepresentability conditions for the Lasso. Extensions to generalized linear models (GLMs) and Cox regression are also discussed. A special case of lasso, “uniReg” is an interesting competitor to good ol’ least squares regression (Legendre, 1805).

This is joint work with Sourav Chatterjee and Trevor Hastie.