BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Estimating the Global Average Treatment Effect under Structured In
 terference - Shuangning Li (University of Chicago\, Booth School of Busine
 ss)
DTSTART:20250313T143000Z
DTEND:20250313T160000Z
UID:TALK229153@talks.cam.ac.uk
CONTACT:Martina Scauda
DESCRIPTION:The field of causal inference develops methods for estimating 
 treatment effects\, often relying on the Stable Unit Treatment Value Assum
 ption (SUTVA)\, which states that a unit’s outcome depends only on its o
 wn treatment. However\, in many real-world settings\, SUTVA is violated du
 e to interference—where the treatment assigned to one unit influences th
 e outcomes of others. Such interference can arise from social interactions
  among units or competition for shared resources\, complicating causal ana
 lysis and leading to biased estimates. Fortunately\, in many cases\, inter
 ference follows structured patterns that can potentially be leveraged for 
 more accurate estimation. In this paper\, we examine and formalize two spe
 cific forms of structured interference—monotone interference and submodu
 lar interference—which we believe arise in many practical settings. We i
 nvestigate how incorporating these structures can improve causal effect es
 timation. Our main contributions are (i) a set of bounds relating key inte
 rference estimands under these structural assumptions and (ii) new estimat
 ors that integrate these structures through constrained optimization. Sinc
 e these constraints may introduce bias\, we further develop debiasing tech
 niques based on treatment regeneration and bootstrap methods to mitigate t
 his issue.\n\nThis is joint work (ongoing) with Kevin Han and Johan Ugande
 r.
LOCATION:MR19\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR