BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Graphs\, Time and Causal Inference
DTSTART:20201030T151500
DTEND:20201030T163000
DTSTAMP:20260407T101355Z
UID:6ca50ea7054184655e0ec1b8238dd9579658c908ea2a96c65c8dd3ce
CATEGORIES:Conferences - Seminars
DESCRIPTION:Vanessa Didelez\, Leibniz Institute for Prevention Research an
 d Epidemiology – BIPS\, Bremen and University of Bremen\nIn this present
 ation\, I will address key aspects of statistical modelling of and causal 
 inference for events in (continuous) time. This is for instance relevant w
 hen some events correspond to some "treatment" and others to important out
 comes such as "relapse" or "death".\n\nFirst we discuss the visual represe
 ntation of multivariate dependence structures among events\, building on a
  marked point processes framework\, using the concept of local independenc
 e and associated graphs (Didelez\, 2008). It will be shown how reasoning a
 nd inference for systems with latent processes can be facilitated using su
 ch graphical representation and a suitable notion of graph-separation\, ca
 lled delta-separation.\n\nSecondly\, I present a formal notion of causal r
 elations between events (or processes) in time based on a decision theoret
 ic approach (Dawid and Didelez\, 2010\; Didelez\, 2015)\, and discuss the 
 use of local independence graphs to decide the question of identifiability
  in systems with unmeasured processes. This will be illustrated with an ap
 plication to cancer screening in Norway (Roysland et al.\, 2020).\n\nMoreo
 ver\, we will address the connection to recent developments in causal medi
 ation and competing events in survival analyses (Didelez\, 2019\; Stensrud
  et al.\, 2020).\n\nThe presentation will focus on basic principles and co
 ncepts rather than technical details.\n \nReferences:\nDawid and Didelez 
 (2010). Identifying the consequences of dynamic treatment strategies: A de
 cision theoretic overview\, Statistics Surveys\, 4\, 184-231.\nDidelez (20
 08). Graphical models for marked point processes based on local independen
 ce. JRSS(B)\, 70\, 245-264.\nDidelez (2015). Causal Reasoning for events i
 n continuous time: a decision-theoretic approach. Proceedings of the 31st 
 Annual Conference on Uncertainty in Artificial Intelligence - Causality Wo
 rkshop (Invited Paper).\nDidelez (2019). Defining causal mediation with a 
 longitudinal mediator and a survival outcome. Lifetime Data Anal 25\, 593
 –610.\nRoysland\, Ryalen\, Nygard\, Didelez (2020). Graphical criteria f
 or identification in continuous-time marginal structural survival models. 
 In preparation.\nStensrud\, Young\, Didelez\, Robins & Hernán (2020) Sepa
 rable Effects for Causal Inference in the Presence of Competing Events\, J
 ASA (online).\n 
LOCATION:zoom https://epfl.zoom.us/j/81101375508
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR