scispace - formally typeset
T

Trevor Hastie

Researcher at Stanford University

Publications -  428
Citations -  230646

Trevor Hastie is an academic researcher from Stanford University. The author has contributed to research in topics: Lasso (statistics) & Feature selection. The author has an hindex of 124, co-authored 412 publications receiving 202592 citations. Previous affiliations of Trevor Hastie include University of Waterloo & University of Toronto.

Papers
More filters
Posted ContentDOI

Fast Numerical Optimization for Genome Sequencing Data in Population Biobanks

TL;DR: In this paper, the authors developed two efficient solvers for optimization problems arising from large-scale regularized regressions on millions of genetic variants sequenced from hundreds of thousands of individuals.
Posted Content

Assessment of Heterogeneous Treatment Effect Estimation Accuracy via Matching

TL;DR: A novel matching distance derived from proximity scores in random forests is introduced and a match‐then‐split principle is proposed for the assessment with cross‐validation of the accuracy of heterogeneous treatment effect estimation.
Posted Content

Feature-weighted elastic net: using "features of features" for better prediction.

TL;DR: The feature-weighted elastic net ("fwelnet") is proposed, which uses these "features of features" to adapt the relative penalties on the feature coefficients in the elastic net penalty to improve prediction.
Posted Content

Nuclear penalized multinomial regression with an application to predicting at bat outcomes in baseball

TL;DR: The authors proposed the nuclear norm penalty as an alternative to the ridge penalty for regularized multinomial regression, which has the advantage of leveraging underlying structure among the response categories to make better predictions.