JSM 2012

A new one-day course, Targeted Learning: Causal Inference for Observational and Experimental Data, was presented at the 2012 Joint Statistical Meetings (JSM) this summer in San Diego, CA on July 29, 2012. The course summary is available on the JSM Web site. The instructors were Mark van der Laan, Maya Petersen, and Sherri Rose.  Download slides from the course here: ZIP file.


HIV Forum 2011

The Forum for Collaborative HIV Research and the University of California, Berkeley School of Public Health presented the short course Statistical Methods for Causal Inference in Observational and Randomized Studies with instructors Mark J. van der Laan, Maya L. Petersen, and Sherri Rose.

This three-day executive-level course in Washington, DC was designed for senior statisticians from regulatory agencies, industry and academic settings and took place September 26-28, 2011.

The course concerned statistical methods for causal inference using observational and experimental longitudinal data. It focused on the application of methodological advances in statistical and causal research to improve the design and interpretation of safety analyses. These analyses will become increasingly important in the post-marketing safety environment for new drugs.

A review and scientific critique of current estimation methods was provided, including an introduction to semi-parametric targeted learning. Structural causal models (causal graphs) and working marginal structural models were introduced as tools for translating a policy question and background knowledge into a target statistical quantity and model. The course emphasized understanding and responding to the challenges posed by safety data in randomized controlled trials as well as observational cohorts, including informative drop out/censoring, missing data, time-dependent confounding, non-compliance, and high dimensional covariates. Examples from the fields of HIV and cardiovascular disease, together with other fields, were used to illustrate the methods and to provide practical experience with analytic design and accurate interpretation of results. The course Web site and course brochure are still available.

Downloads

Outline: [pdf]

Slides: Day 1–Part 1 [zip], Day 1–Part 2 [zip], Day 2 [zip], Day 3 [zip]

Brief R package demos: “SuperLearner” [.txt] and “tmle” [.txt]