Sample Size Determination in Bayesian Clinical Trials with Clustered Data / Positive and Unlabeled Data: Model, Estimation, Inference, and Classification
Luke Hagar, PhD
Postdoctoral Scholar
Department of Epidemiology, Biostatistics and Occupational Health | 9I制作厂免费
Chi-Kuang Yeh, PhD
Postdoctoral Scholar
9I制作厂免费 & University of Waterloo
WHEN:听Wednesday, February 19, 2025, from 3:30 to 4:30 p.m.
WHERE:听Hybrid | 2001 9I制作厂免费 College Avenue, Room 1140;
NOTE:听Luke Hagar & Chi-Kuang Yeh will be presenting in-person
Abstract
Sample Size Determination in Bayesian Clinical Trials with Clustered Data, Luke Hagar, PhD:听When designing Bayesian clinical trials, operating characteristics are typically assessed by estimating the sampling distribution of posterior summaries via Monte Carlo simulation. This process is computationally intensive, particularly for trials with clustered data. We propose an efficient method to assess operating characteristics and determine sample sizes for Bayesian trials with clustered data and multiple endpoints. We prove theoretical results that enable posterior probabilities to be modelled as a function of the sample size. Using these functions, we assess operating characteristics at a range of sample sizes given simulations conducted at only two sample sizes. Our methodology is illustrated using a current clinical trial with clustered data.
Positive and Unlabeled Data: Model, Estimation, Inference, and Classification,听Chi-Kuang Yeh, PhD:听Case-control is a study design widely used in biomedical research to investigate the causes of diseases. However, data contamination is a common issue in case-control studies due to, for instance, some medical conditions may go unrecognized in many patients, and they are misclassified as healthy one. This situation may be characterized as positive and unlabeled (PU) data. We introduce new approach to addressing through the double exponential tilting model (DETM). Traditional methods often fall short because they only apply to selected completely at random PU data, where the labeled positive and unlabeled positive data are assumed to be from the same distribution. In contrast, our DETM's dual structure effectively accommodates the more complex and underexplored selected at random PU data, where the labeled and unlabeled positive data can be from different distributions. Through theoretical insights and practical applications, this study highlights DETM as a comprehensive framework for addressing the challenges of PU data.
Speaker Bio
Luke Hagar is a Postdoctoral Scholar in the Department of Epidemiology, Biostatistics and Occupational Health at 9I制作厂免费, supervised by Dr. Shirin Golchi. His research leverages theory to make simulation-based methods for experimental design more economical. For more information, please visit website:
Chi-Kuang Yeh received his Ph.D. in Statistics from the University of Waterloo in 2023. He is currently a joint postdoc at 9I制作厂免费 and the University of Waterloo sponsor by the Canadian Statistical Science Institute. His research area include functional data analysis, statistical machine learning, dependence modeling, and optimal experimental design. For more information, please visit website: .