BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:When the Causal Graph is Unknown\, Embrace Model Pluralism: Some I
 deas and Challenges - Rohit Bhattacharya (Williams College)
DTSTART:20260304T091500Z
DTEND:20260304T100000Z
UID:TALK244387@talks.cam.ac.uk
DESCRIPTION:A fundamental challenge in observational causal inference is t
 he correct specification of a causal model. When identification assumption
 s are uncertain\, analysts may seek to triangulate effect estimates across
  multiple candidate models that rely on different assumptions. Yet princip
 led methods for quantitative triangulation remain underdeveloped. We devel
 op two complementary frameworks to address this gap. The first concerns te
 sting the causal null hypothesis and yields a procedure that is valid when
 ever at least one candidate model is correctly specified. The second focus
 es on effect estimation and produces a combined estimator that is consiste
 nt provided at least one model is both correct and empirically testable. T
 ogether\, these frameworks formalize robustness under methodological plura
 lism without relying on model selection or agreement across models&mdash\;
 an assumption often implicit in the triangulation literature. Throughout t
 he talk\, I will also weave in discussions on theoretical and practical ch
 allenges that arose (and some that still remain) in implementing such fram
 eworks.\nCo-authors on these works: Ted Westling\, Youjin Lee\, Junhui Yan
 g\, He Bai\, Ina Ocelli
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR