Online Program Home
My Program

Abstract Details

Activity Number: 43 - SPEED: Statistics in Sports; Physical Activity/Sleep Studies, and Nonparametrics Part 1
Type: Contributed
Date/Time: Sunday, July 28, 2019 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #307225
Title: Weighted Regression with Covariates Derived from Discrepancies Between High-Dimensional Predictors
Author(s): Lucia Tabacu* and Andrew Leroux and Ciprian Crainiceanu
Companies: Old Dominion University and JHU and Johns Hopkins University
Keywords: Accelerometry; NHANES; Survey weights; Physical activity; Hamming distance; Nonparametric regression

We consider the problem of predicting scalar outcomes based on mixed type (continuous or discrete) high dimensional predictors and scalar covariates that account for survey sample weights. The idea is to build a small number of predictors based on the discrepancy between these high dimensionality predictors and a reference set. Moreover, the approach allows one to transform the difficult problem of choice of distance into a standard covariate selection problem and allows the incorporation of survey sampling weights. The methods are motivated by a study of the association between 5-year all-cause mortality in the National Health and Nutrition Examination Survey (NHANES) 2003-2006 waves and physical activity measured at the minute level using accelerometers.

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

Back to the full JSM 2019 program