The Fourth Annual Atlantic Causal Inference Conference (ACIC) Data Challenge provides an opportunity to compare causal inference methodologies across a variety of data generating processes (DGP).
As in previous years, the challenge focuses on computational methods of inferring causal effects from quasi-real world data. This year’s target of estimation is the population average treatment effect (ATE). There will be two tracks:
- Low dimensional datasets (varying size, e.g., 500 x 20)
- High dimensional datasets (varying size, e.g. 1000 x 200, 2000 x 200)
Participants can download datasets (between 2000 and 3000 datasets in each track), run analyses using their own computing resources, and upload results to the Challenge website for evaluation.
Datasets are available for download here (no registration required) (bottom of the page).
The deadline for submitting results is April 15, 2019.
The 2019 Data Challenge is now open. Preliminary results will be announced during the conference.
- mid-December, 2018: The Challenge website goes live. Sample datasets that can be used to develop your approach will be available for download.
- mid-January, 2019: Challenge Datasets available for download
- mid-April, 2019: Deadline for results files to be uploaded
Susan Gruber, Putnam Data Sciences, LLC
Geneviève Lefebvre, Université du Québec à Montréal
Tibor Schuster, McGill University
Alexandre Piche, MILA, Université de Montréal, Element AI
For more Information, contact sgruber [at] putnamds.com (subject: Data%20Challenge%20ACIC%202019) (Susan Gruber)