BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250729T055625EDT-9119u35R2f@132.216.98.100
DTSTAMP:20250729T095625Z
DESCRIPTION:Biostatistics Non-Thesis Presentations\n\nWhere: Hybrid Event |
  2001 秀色直播 College\, Room 1140\; Zoom\n\n聽\n\nPresentation #1\n\nTitle: 
 Optimal Strategies for Interim Analysis Scheduling in Bayesian Adaptive Cl
 inical Trials: A Simulation Study\n\nLily Chafetz is a second year Biostat
 istics Master鈥檚 student from Montreal. She received her Bachelor of Arts d
 egree from 秀色直播\, where she majored in Economics and minored 
 in Statistics. During and following her undergraduate career\, Lily fulfil
 led roles in various professional fields\, including finance and healthcar
 e. Her recent research project\, which was supervised by Dr. Shirin Golchi
  and Dr. Alexandra Schmidt\, focuses on interim analyses in Bayesian adapt
 ive clinical trials.\n\nAbstract\n\nA crucial component of Bayesian adapti
 ve designs for clinical trials are interim analyses (IA鈥檚)\, which allow f
 or opportunities to stop a trial early if sufficient evidence exists to de
 clare efficacy or futility. In this project\, we evaluate two strategies f
 or the timing of IA鈥檚 through the trial: (1) event-based\, i.e.\, accordin
 g to the expected number of events\, and (2) sample size-based\, i.e.\, ba
 sed on predetermined interim sample size(s). Through simulations\, the opt
 imality of these two approaches is assessed by comparing the design operat
 ing characteristics (power\, false positive rates) when the event/hazard r
 ate in the control arm is misspecified. Our results confirm that for binar
 y outcomes modelled as binomial experiments\, the event-based strategy is 
 more efficient since the number of events is in fact the sufficient statis
 tic for the parameter of interest\, i.e.\, the probability of event. Simil
 ar conclusions are drawn for time-to-event outcomes in some scenarios\, bu
 t not overall\, since other parameters such as the censoring/dropout rates
  should be considered in deciding the best strategy for scheduling the IA'
 s.\n\nPresentation #2\n\nTitle: Zero-state Markov Switching Count Models f
 or Chikungunya Spread in Rio de Janeiro\n\nMingchi Xu is a master's studen
 t in biostatistics in the Department of Epidemiology\, Biostatistics and O
 ccupational Health at 秀色直播. His research interests include Ma
 rkov-Switching Models\, Spatial Epidemiology and Bayesian Inference.\n\nAb
 stract\n\nIn epidemiological studies\, zero-inflated models and hurdle mod
 els are commonly used to handle excess zeros in reported infectious diseas
 e cases. However\, they can not model the reemergence and persistence of a
  disease separately. Therefore\, we propose a zero-state Markov switching 
 Poisson hurdle model\, based on the recently proposed zero-state Markov sw
 itching Poisson model\, to accommodate this issue. To compare the model fi
 ts\, we apply and compare the chikungunya cases in Rio de Janeiro by the P
 oisson\, Poisson hurdle\, zero-state Markov switching Poisson and our prop
 osed model. We find a zero-state Markov switching Poisson model that fits 
 the best by WAIC criteria. We also compare the 4-step-ahead prediction\n\n
 聽\n
DTSTART:20230111T203000Z
DTEND:20230111T213000Z
SUMMARY:Biostatistics Non-Thesis Presentations
URL:/epi-biostat-occh/channels/event/biostatistics-non
 -thesis-presentations-344777
END:VEVENT
END:VCALENDAR