BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20250725T031518EDT-198416kuuR@132.216.98.100 DTSTAMP:20250725T071518Z DESCRIPTION:Linbo Wang\, PhD\n\nAssistant Professor\n Department of Statisti cal Sciences | University of Toronto\n\nWHEN: Wednesday\, November 15\, 20 23\, from 3:30 to 4:30 p.m.\n\nWHERE: Hybrid | 2001 ºÚÁϲ»´òìÈ College\, Rm 11 40 | Zoom &\n\nNote: Dr. Wang will present from Toronto\n\nAbstract\n\nIn many observational studies\, researchers are often interested in studying the effects of multiple exposures on a single outcome. Standard approaches for high-dimensional data such as the lasso assume the associations betwe en the exposures and the outcome are sparse. These methods\, however\, do not estimate the causal effects in the presence of unmeasured confounding. In this paper\, we consider an alternative approach that assumes the caus al effects in view are sparse. We show that with sparse causation\, the ca usal effects are identifiable even with unmeasured confounding. At the cor e of our proposal is a novel device\, called the synthetic instrument\, th at in contrast to standard instrumental variables\, can be constructed usi ng the observed exposures directly. We show that under linear structural e quation models\, the problem of causal effect estimation can be formulated as an â„“0-penalization problem\, and hence can be solved efficiently using off-the-shelf software. Simulations show that our approach outperforms st ate-of-art methods in both low-dimensional and high-dimensional settings. We further illustrate our method using a mouse obesity dataset.\n\nSpeaker Bio\n\nhttps://sites.google.com/site/linbowangpku/home\n DTSTART:20231115T203000Z DTEND:20231115T213000Z SUMMARY:The synthetic instrument: From sparse association to sparse causat ion URL:/epi-biostat-occh/channels/event/synthetic-instrum ent-sparse-association-sparse-causation-352629 END:VEVENT END:VCALENDAR