BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260924T150958EDT-2214wVnvCR@132.216.98.100
DTSTAMP:20260924T190958Z
DESCRIPTION:\n	\n		\n			\n				Dehan Kong\, PhD\n\n				Associate Professor in Statistics\,
  University of Toronto\n\n				Note: Meet & Greet Prof Dehan Kong from 3-3:30pm
  in Room 1140\; Prior to seminar 3:30-4:30pm\n\n				WHEN: Wednesday\, Septembe
 r 10\, 2025\, from 3:30 to 4:30 p.m.\n					WHERE: Hybrid | 2001 McGill College 
 Avenue\, Rm 1140\; Zoom\n					NOTE: Dehan Kong will be presenting in-person\n\n
 				Abstract\n\n				Instrumental variable methods provide useful tools for inferri
 ng causal effects in the presence of unmeasured confounding. To apply thes
 e methods with large-scale data sets\, a major challenge is to find valid 
 instruments from a possibly large candidate set. In practice\, most of the
  candidate instruments are often not relevant for studying a particular ex
 posure of interest. Moreover\, not all relevant candidate instruments are 
 valid as they may directly influence the outcome of interest. In this arti
 cle\, we propose a data-driven method for causal inference with many candi
 date instruments that addresses these two challenges simultaneously. A key
  component of our proposal involves using pseudo variables\, known to be i
 rrelevant\, to remove variables from the original set that exhibit spuriou
 s correlations with the exposure. Synthetic data analyses show that the pr
 oposed method performs favourably compared to existing methods. We apply o
 ur method to a Mendelian randomization study estimating the effect of obes
 ity on health-related quality of life.\n\n				Speaker Bio\n\n				I am currently an
  associate professor in statistics at the University of Toronto. I receive
 d my B.S. in Mathematics from Nankai University in 2008\, and my Ph.D. in 
 Statistics from North Carolina State University in 2013. I was a postdocto
 ral fellow in the Department of Biostatistics at the University of North C
 arolina\, Chapel Hill from 2013-2016. My research aims to develop advanced
  data science tools and methodologies to handle large\, complex\, multi-sc
 ale real-world data. I work on topics including statistical machine learni
 ng\, neuroimaging data analysis\, statistical genetics and genomics\, and 
 causal inference. My research is being supported by the Natural Sciences a
 nd Engineering Research Council of Canada (NSERC)\, the Canadian Institute
 s of Health Research (CIHR)\, the University of Toronto’s Data Science Ins
 titute\, Canadian Statistical Sciences Institute (CANSSI)\, CANSSI Ontario
 \, and Mitacs.\n\n				Dehan Kong's Website\n\n				 \n			\n		\n	\n\n
DTSTART:20250910T193000Z
DTEND:20250910T203000Z
SUMMARY:Fighting Noise with Noise: Causal Inference with Many Candidate Ins
 truments
URL:https://www.mcgill.ca/epi-biostat-occh/channels/event/fighting-noise-no
 ise-causal-inference-many-candidate-instruments-367095
END:VEVENT
END:VCALENDAR