BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250510T211559EDT-9412pvMaZ4@132.216.98.100
DTSTAMP:20250511T011559Z
DESCRIPTION:Luke Hagar\, PhD\n\nPostdoctoral Scholar\n	Department of Epidemi
 ology\, Biostatistics and Occupational Health | 涩里番\n\nChi-K
 uang Yeh\, PhD\n\nPostdoctoral Scholar\n	涩里番 & University of 
 Waterloo\n\nWHEN:聽Wednesday\, February 19\, 2025\, from 3:30 to 4:30 p.m.
 \n	WHERE:聽Hybrid | 2001 涩里番 College Avenue\, Room 1140\; Zoom\n	NOTE:聽Luk
 e Hagar & Chi-Kuang Yeh will be presenting in-person\n\nAbstract\n\nSample
  Size Determination in Bayesian Clinical Trials with Clustered Data\, Luke
  Hagar\, PhD:聽When designing Bayesian clinical trials\, operating characte
 ristics are typically assessed by estimating the sampling distribution of 
 posterior summaries via Monte Carlo simulation. This process is computatio
 nally intensive\, particularly for trials with clustered data. We propose 
 an efficient method to assess operating characteristics and determine samp
 le sizes for Bayesian trials with clustered data and multiple endpoints. W
 e prove theoretical results that enable posterior probabilities to be mode
 lled as a function of the sample size. Using these functions\, we assess o
 perating characteristics at a range of sample sizes given simulations cond
 ucted at only two sample sizes. Our methodology is illustrated using a cur
 rent clinical trial with clustered data.\n\nPositive and Unlabeled Data: M
 odel\, Estimation\, Inference\, and Classification\,聽Chi-Kuang Yeh\, PhD:聽
 Case-control is a study design widely used in biomedical research to inves
 tigate the causes of diseases. However\, data contamination is a common is
 sue in case-control studies due to\, for instance\, some medical condition
 s may go unrecognized in many patients\, and they are misclassified as hea
 lthy one. This situation may be characterized as positive and unlabeled (P
 U) data. We introduce new approach to addressing through the double expone
 ntial tilting model (DETM). Traditional methods often fall short because t
 hey only apply to selected completely at random PU data\, where the labele
 d positive and unlabeled positive data are assumed to be from the same dis
 tribution. In contrast\, our DETM's dual structure effectively accommodate
 s the more complex and underexplored selected at random PU data\, where th
 e labeled and unlabeled positive data can be from different distributions.
  Through theoretical insights and practical applications\, this study high
 lights DETM as a comprehensive framework for addressing the challenges of 
 PU data.\n\nSpeaker Bio\n\nLuke Hagar is a Postdoctoral Scholar in the Dep
 artment of Epidemiology\, Biostatistics and Occupational Health at 涩里番 
 University\, supervised by Dr. Shirin Golchi. His research leverages theor
 y to make simulation-based methods for experimental design more economical
 . For more information\, please visit website: https://lmhagar.github.io/.
 \n	\n	Chi-Kuang Yeh received his Ph.D. in Statistics from the University of 
 Waterloo in 2023. He is currently a joint postdoc at 涩里番 and
  the University of Waterloo sponsor by the Canadian Statistical Science In
 stitute. His research area include functional data analysis\, statistical 
 machine learning\, dependence modeling\, and optimal experimental design. 
 For more information\, please visit website: https://chikuang.github.io/.
 \n
DTSTART:20250219T203000Z
DTEND:20250219T213000Z
SUMMARY:Sample Size Determination in Bayesian Clinical Trials with Clustere
 d Data / Positive and Unlabeled Data: Model\, Estimation\, Inference\, and
  Classification
URL:/epi-biostat-occh/channels/event/sample-size-deter
 mination-bayesian-clinical-trials-clustered-data-positive-and-unlabeled-da
 ta-model-363058
END:VEVENT
END:VCALENDAR