|
Activity Number:
|
178
- Recent Development on the Analysis of Time-to-Event Data
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
|
Abstract #312234
|
|
|
Title:
|
Dual Response Surface Methodology for Survival Data
|
|
Author(s):
|
Haiyan Yu* and Muzi Zhang and Dennis K.J. Lin and Mingyue Gao
|
|
Companies:
|
Penn State University and and The Pennsylvania State University, University Park and University College London
|
|
Keywords:
|
Balancing score;
Covariance models;
Factorial designs;
Interquartile range;
Response surface methodology;
Survival data
|
|
Abstract:
|
In many experiments, not only the primary response is used to build the model, but also the secondary response is applied to capture the dispersion of sample means. Despite the controllable inputs of treatment, covariates as environmental inputs are used to build the metamodels for separating the blocks in experiments. The environmental inputs in the model also bring block effects for dual responses. Different blocks may take a different optimal assignment of treatment to yield outcomes. Our approach provides a weighted solution to such an assignment task in the dual-response problem with survival data. This solution enables us to choose the optimal treatment for each block after observing the values of covariates. The balancing score is introduced as a concomitant variable to split the blocks of units. The primary response and secondary response are acquired from the median and the interquartile range of the distribution of survival probabilities. The theoretical results guarantee the identifiability of the solutions with distributively robust optimal values. Examples of survival data are applied to validate the efficacy of this method.
|
Authors who are presenting talks have a * after their name.
