Activity Number:
|
339
- SPEED: Biopharmaceutical and General Health Studies: Statistical Methods and Applications, Part 1
|
Type:
|
Contributed
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
|
Sponsor:
|
ENAR
|
Abstract #307391
|
Presentation 1
Presentation 2
Presentation 3
Presentation 4
|
Title:
|
Estimating the Relative Risk for Response-Biased Samples: Calibration and Conditional Likelihood
|
Author(s):
|
Claudia Rivera-Rodriguez*
|
Companies:
|
University of Auckland
|
Keywords:
|
|
Abstract:
|
Relative risks are a popular measure of treatment effect for binary outcomes. Methods for estimation of the relative ris have been evaluated and implemented in settings when the sample is random, but not when the sample arises from a two-phase design. These designs s are used in the analysis of missing data or when some information can only be collected on a subset of individuals. We focus on statistical inference for the relative risk under two-phase designs. Methods implemented include weighted likelihood with calibrated weights and conditional likelihood with estimated weights. The results are valid for general link functions or distributions, but we exemplify it with the log-model.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2019 program
|