BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian regression discontinuity design with unknown cut-off - St
 ephanie van der Pas (Vrije Universiteit Amsterdam)
DTSTART:20250509T104500Z
DTEND:20250509T114500Z
UID:TALK230506@talks.cam.ac.uk
DESCRIPTION:Joint work with Julia Kowalska and Mark van de Wiel. Regressio
 n discontinuity design (RDD) is a quasi-experimental approach used to esti
 mate the causal effects of an intervention assigned based on a cutoff crit
 erion. RDD exploits the idea that close to the cutoff units below and abov
 e are similar\; hence\, they can be meaningfully compared. Consequently\, 
 the causal effect can be estimated only locally at the cutoff point. This 
 makes the cutoff point an essential element of RDD. However\, especially i
 n medical applications\, the exact cutoff location may not always be discl
 osed to the researcher\, and even when it is\, the actual location may dev
 iate from the official one. As we illustrate on the application of RDD to 
 the HIV treatment eligibility data\, estimating the causal effect at an in
 correct cutoff point leads to meaningless results. Moreover\, since the cu
 toff criterion often acts as a guideline rather than as a strict rule\, th
 e location of the cutoff may be unclear from the data. The method we prese
 nt can be applied both as an estimation and validation tool in RDD. We use
  a Bayesian approach to incorporate prior knowledge and uncertainty about 
 the cutoff location in the causal effect estimation. At the same time\, ou
 r Bayesian model LoTTA is fitted globally to the whole data\, whereas RDD 
 is a local\, boundary point estimation problem. In this work we address a 
 natural question that arises: how to make Bayesian inference more local to
  render a meaningful and powerful estimate of the treatment effect?
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR