Joint Initiative for Causal Inference (JICI) Webinar Series - Spring 2021

Joint Initiative for Causal Inference Webinar Series

Panel Discussion

March 3, 2021
7:00am to 9:00am
Virtual Participation


The Joint Initiative for Causal Inference (JICI) Webinar Series was launched on March 3, 2021, and is hosted by Drs. Mark van der Laan and BIDS Faculty Affiliate Maya Petersen of Berkeley Public Health’s Center for Targeted Machine Learning (CTML), and featuring speakers from the University of Copenhagen and Novo Nordisk, a leading global healthcare company headquartered in Denmark. Sessions will be held from 7:00—9:00 AM PT on the first Wednesdays of the month (March 3rd, April 7th, May 5th, and June 2nd)  with presentations on utilizing causal inference and targeted learning methods to answer pressing health questions in the modern methodological and data ecosystem. Participants can attend all days, just join for individual sessions, or watch recordings of the sessions on the event site. Register for these free events. 

See the videos and slides from Session 1 on March 3, 2021, in which host Maya Petersen — BIDS Faculty Affilate and Chair of the Division of Biostatistics and Associate Professor Epidemiology and Biostatistics in Berkeley's School of Public Health — provided the introduction to the webinar series, and then Søren Rasmussen and Thomas Gerds introduced the challenges posed by competing risks from both an industry and academic perspective. David Chen then presented on current methods and problems in handling competing risks, including discrete-time targeted maximum likelihood estimation results from simulations and analysis of the LEADER trial.  Helene Rytgaard then provided a general introduction to novel methods for competing risks: continuous time and 1-step targeted maximum likelihood estimation, as well as applications. There was also an open Q&A session for all presentations at the end of the webinar.

  1. Competing risk in clinical trials – a brief viewSøren Rasmussen, Novo Nordisk
  2. Competing risks in medical research, Thomas Alexander Gerds, University of Copenhagen
  3. Analyzing Competing Risks with Discrete Time TMLEDavid Chen, UC Berkeley
  4. Continuous-time one-step TMLE for competing risksmHelene Rytgaard, University of Copenhagen

Contact: Questions about this series may be addressed to Lucas Carlton ( and


Maya Petersen

Associate Professor, Divisions of Biostatistics and Epidemiology, School of Public Health, UC Berkeley

Maya Petersen is Associate Professor of Biostatistics and Chair of the Division of Biostatistics at the UC Berkeley School of Public Health. Dr. Petersen’s methodological research focuses on the development and application of novel causal inference methods to problems in health, with an emphasis on longitudinal data and adaptive treatment strategies (dynamic regimes), machine learning methods, and study design and analytic strategies for impact evaluation. Dr. Petersen’s applied work focuses on developing and evaluating improved HIV prevention and care strategies in resource-limited settings. She serves as MPI (with Dr. Diane Havlir and Dr. Moses Kamya) for the Sustainable East Africa Research in Community Health (SEARCH) consortium, and co-leads (with Profs. Mark van der Lan and Alan Hubbard) the Berkeley SPH Center for Targeted Learning.