|
Activity Number:
|
42
- Novel Statistical Methods with a Biostatistics Leaning
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 28, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
ENAR
|
|
Abstract #307167
|
Presentation
|
|
Title:
|
Nonparametric Conditional Density Estimation for Pooled Biomarker Data
|
|
Author(s):
|
Dewei Wang* and Xichen Hou and Joshua Tebbs
|
|
Companies:
|
University of South Carolina and University of South Carolina and University of South Carolina
|
|
Keywords:
|
Nonparametric deconvoluation;
Density estimation;
Pooled data
|
|
Abstract:
|
In biomarker studies, when resources (e.g., the budget and/or the number of specimens) are limited, pooling specimens to take measurements of the biomarker's concentration level is often an alternative means. This article develops a kernel-based regression estimator of a biomarker level's density when a continuous covariate is available for each specimen but the biomarker is measured in pools with measurement errors. Consistency and asymptotic normality of our estimator is established. The rates of convergence depend on the tail behavior of the characteristic functions of the measurement error and the biomarker level. Simulation studies demonstrate the practical advantages of our method when comparing to the existing work. We further illustrate our method via a Polyfluorochemical data set.
|
Authors who are presenting talks have a * after their name.
