Online Program Home
My Program

Abstract Details

Activity Number: 8 - Machine Learning Methods and Applications: Making an Impact in Biomedical Research
Type: Invited
Date/Time: Sunday, July 28, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistical Learning and Data Science
Abstract #300273 Presentation
Title: RELIEF-Based Feature Selection for Heterogeneous Treatment Effects with Massive Data
Author(s): Xiaogang Su*
Companies: University of Texas, El Paso
Keywords: Feature Selection; Heterogeneous Treatment Effects; Precision medicine; RELIEF

We extend the RELIEF machine learning techniques for feature selection in estimating heterogeneous treatment effects with big data. The proposed method facilitates a variable importance ranking for covariates in terms of their predictive (vs. prognostic) values in modifying the treatment effects. The method can also be oriented towards the assessment of qualitative treatment-by-covariate interactions. Properties of the proposed method are explored, including a statistical interpretation in association with the K nearest neighbors (KNN). We demonstrate the usefulness and efficiency of the proposed method and make comparison with other competitive approaches via both simulation and real data applications.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program