BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Causal Identification: Selection Bias and Latent Variables
DTSTART:20230821T150000
DTEND:20230821T170000
DTSTAMP:20260511T212705Z
UID:08444d3826d73a36d9680ef4145e891156c8ad6a3d49cea3bc682d1b
CATEGORIES:Conferences - Seminars
DESCRIPTION:Amir Aboueimehrizi\nEDIC candidacy exam\nExam president: Prof.
  Lenka Zdeborová\nThesis advisor: Prof. Negar Kiyavash\nThesis co-advisor
 : Prof. Matthias Grossglauser\nCo-examiner: Prof. Yanina Shkel\n\nAbstract
 \nCausal inference aims to identify the effects of some\nvariables on othe
 rs within a system. Depending on the objective\nand available data\, diver
 se scenarios can emerge. This report\npresents an overview of three distin
 ct problems: the identification\nof causal effects (1) on the entire popul
 ation from population data\,\n(2) on a specific sub-population from observ
 ational data of the\npopulation\; and (3) on the entire population from ob
 servational\ndata of only a sub-population\, where unobservable variables 
 might\nexist in all setting.\nFurthermore\, we discuss our ongoing researc
 h\, identifying\ncausal effects on a sub-population using the observationa
 l\ndistribution of that sub-population.\n\nBackground papers\n1- Identifi
 cation of Joint Interventional Distributions in Recursive Semi-Markovian C
 ausal Models\n2- Identification of Conditional Interventional Distributio
 ns\n3- Recovering Causal Effects from Selection Bias\n 
LOCATION:BC 129 https://plan.epfl.ch/?room==BC%20129
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR