Colloquium
- Home
- About
- Academics
- Courses
- Undergraduate
- Graduate
- Alumni
- Career Guidance
- Giving
- Research
- Resources
- Stat Consulting Center
- Directory
- Contact Us
Statistics Colloquium
September 13, 11 a.m.
Alavi Commons Room, 6625 Everett Tower
Duy Ngo
Department of Statistics
Abstract:
The importance of demonstrating patient subgroups who benefit from a treatment in a clinical trial is increasingly recognized by regulators and health technology assessment agencies worldwide. Most proposed methods that currently addressed this issue only focus on treatment-covariate interactions and do not fully account for multiplicity. To overcome these limitations, we introduce the Bayesian credible subgroups survival analysis method to identify the baseline covariate profiles of patients who benefit from treatment. We estimate the treatment effect within the identified benefiting subgroup using log hazard ratio and restricted mean survival time (RMST). The advantages of our approach are that: (1) it does not require pre-specification of subgroups and work directly with the covariate space, and (2) it naturally makes statistical inferences from the full posterior distribution. We also investigate frequentist properties of this method and compare it to other methods such as Bayesian regression tree in a simulation study. In general, the nominal coverage has been preserved, however at lower sensitivity in some scenarios. Our methods applied to a case study of prostate carcinoma.
Bio:
Duy Ngo is Assistant Professor in the Department of Statistics at Western Michigan University. He received his Ph.D. from the University of California, Irvine. His research interests include Bayesian analysis, clustering, time series, spectral analysis, and functional data analysis. He is also working on statistical modeling and machine learning for subgroup identification/analysis for benefit–risk assessment.
---
All statistics graduate students are expected to attend.