Online Causal Inference Seminar
A regular international causal inference seminar.
Upcoming Seminars
All seminars are on Tuesdays at 8:30 am PT / 11:30 am ET / 4:30 pm London / 5:30 pm Berlin / 11:30 pm Beijing.
You can join the webinar on Zoom here (webinar ID: 968 8371 7451). The password is 414559.
Tuesday, October 29, 2024: Toru Kitagawa (Brown University)
- Title: Policy Choice in Time-Series by Empirical Welfare Maximization
- Discussant: Mikkel Plagborg-Moller (Princeton University)
- Abstract: This paper develops a novel method for policy choice in a dynamic setting where the available data is a multivariate time series. Building on the statistical treatment choice framework, we propose Time-series Empirical Welfare Maximization (T-EWM) methods to estimate an optimal policy rule for the current period or over multiple periods by maximizing an empirical welfare criterion constructed using nonparametric potential outcome time-series. We characterize conditions under which T-EWM consistently learns a policy choice that is optimal in terms of conditional welfare given the time-series history. We then derive a nonasymptotic upper bound for conditional welfare regret and its minimax lower bound. To illustrate the implementation and uses of T-EWM, we perform simulation studies and apply the method to estimate optimal monetary policy rules from macroeconomic time-series data.
[Paper]Tuesday, November 5, 2024 (Young researcher seminar)
Speaker 1: Jinzhou Li (Stanford University)
- Title: Root cause discovery via permutations and Cholesky decomposition
- Abstract: This work is motivated by the following problem: Can we identify the disease-causing gene in a patient affected by a monogenic disorder? This problem is an instance of root cause discovery. In particular, we aim to identify the intervened variable in one interventional sample using a set of observational samples as reference. We consider a linear structural equation model where the causal ordering is unknown. We begin by examining a simple method that uses squared z-scores and characterize the conditions under which this method succeeds and fails, showing that it generally cannot identify the root cause. We then prove, without additional assumptions, that the root cause is identifiable even if the causal ordering is not. Two key ingredients of this identifiability result are the use of permutations and the Cholesky decomposition, which allow us to exploit an invariant property across different permutations to discover the root cause. Furthermore, we characterize permutations that yield the correct root cause and, based on this, propose a valid method for root cause discovery. We also adapt this approach to high-dimensional settings. Finally, we evaluate the performance of our methods through simulations and apply the high-dimensional method to discover disease-causing genes in the gene expression dataset that motivates this work.
[Paper]
Speaker 2: Yuyao Wang (University of California San Diego)
- Title: Learning treatment effects under covariate dependent left truncation and right censoring
- Abstract: In observational studies with delayed entry, causal inference for time-to-event outcomes can be challenging. The challenges arise because, in addition to the potential confounding bias from observational data, the collected data often also suffers from the selection bias due to left truncation, where only subjects with time to event (such as death) greater than the enrollment times are included, as well as bias from informative right censoring. To estimate the treatment effects on time-to-event outcomes in such settings, inverse probability weighting (IPW) is often employed. However, IPW is sensitive to model misspecifications, which makes it vulnerable, especially when faced with three sources of biases. Moreover, IPW is inefficient. To address these challenges, we propose a doubly robust framework to handle covariate dependent left truncation and right censoring that can be applied to a wide range of estimation problems, including estimating average treatment effect (ATE) and conditional average treatment effect (CATE). For average treatment effect, we develop estimators that enjoy model double robustness and rate double robustness. For conditional average treatment effect, we develop orthogonal and doubly robust learners that can achieve oracle rate of convergence. Our framework represents the first attempt to construct doubly robust estimators and orthogonal learners for treatment effects that account for all three sources of biases: confounding, selection from covariate-induced dependent left truncation, and informative right censoring.Tuesday, November 19, 2024: Jared S. Murray (McCombs School of Business)
- Title: TBA
- Discussant: TBA
- Abstract: TBATuesday, December 03, 2024: Yiqing Xu (Stanford University)
- Title: Factorial Difference-in-Differences
- Discussant: TBA
- Abstract: In many social science applications, researchers use the difference-in-differences (DID) estimator to establish causal relationships, exploiting cross-sectional variation in a baseline factor and temporal variation in exposure to an event that presumably may affect all units. This approach, which we term factorial DID (FDID), differs from canonical DID in that it lacks a clean control group unexposed to the event after the event occurs. In this paper, we clarify FDID as a research design in terms of its data structure, feasible estimands, and identifying assumptions that allow the DID estimator to recover these estimands. We frame FDID as a factorial design with two factors: the baseline factor, denoted by G, and the exposure level to the event, denoted by Z, and define the effect modification and causal interaction as the associative and causal effects of G on the effect of Z, respectively. We show that under the canonical no anticipation and parallel trends assumptions, the DID estimator identifies only the effect modification of G in FDID, and propose an additional factorial parallel trends assumption to identify the causal interaction. Moreover, we show that the canonical DID research design can be reframed as a special case of the FDID research design with an additional exclusion restriction assumption, thereby reconciling the two approaches. We extend this framework to allow conditionally valid parallel trends assumptions and multiple time periods, and clarify assumptions required to justify regression analysis under FDID. We illustrate these findings with empirical examples from economics and political science, and provide recommendations for improving practice and interpretation under FDID.
This is joint work with Anqi Zhao and Peng Ding.
[Paper]Tuesday, December 10, 2024: Panos Toulis and Wenxuan Guo (University of Chicago)
- Title: TBA
- Discussant: TBA
- Abstract: TBA
Mailing List
To stay up-to-date about upcoming presentations and receive Zoom invitations please join our mailing list. You will receive an email to confirm your subscription. If you are already subscribed to our mailing list and would like to unsubscribe, you may unsubcribe here.
Suggest a speaker
If there is anyone you would like to hear at the Online Causal Inference Seminar, you may let us know here.
Opportunities in Causal Inference
Please check out our opportunities in causal inference page for conferences, workshops, and job listings! If you would like us to list an opportunity, please email us at onlinecausalinferenceseminar@gmail.com.
Youtube channel
Follow us on youtube!
Format and Rules
The seminars are held on Zoom and last 60 minutes. Our seminars will typically follow one of three formats:
Format 1: single presentation
45 minutes of presentation
10 minutes of discussion, led by an invited discussant
Q&A, time permitting
Format 2: two presentations
Two presentations, 25-30 minutes each
Q&A, time permitting
Format 3: interview
40-45 minute conversation with leader in causal inference
15-20 minutes of Q&A
A moderator collects audience questions in Q&A section.
Moderators may ask you to unmute yourself to participate in the discussion. Please note that you may be recorded if you activate your audio or video during the seminar.
Organizers & Moderators
Naoki Egami (Columbia), Laura Forastiere (Yale), Guido Imbens (Stanford), Ying Jin (Stanford), Sara Magliacane (University of Amsterdam), Razieh Nabi (Emory), Georgia Papadogeorgou (University of Florida), Ema Perkovic (UWashington), Dominik Rothenhäusler (Stanford), Qingyuan Zhao (Cambridge), Michael Celentano (Stanford)
Advising committee
Susan Athey (Stanford), Guillaume Basse (Stanford), Peter Bühlmann (ETH Zürich), Peng Ding (Berkeley), Andrew Gelman (Columbia), Guido Imbens (Stanford), Fabrizia Mealli (Florence), Nicolai Meinshausen (ETH Zürich), Maya Petersen (Berkeley), Thomas Richardson (UW), Dominik Rothenhäusler (Stanford), Jas Sekhon (Berkeley/Yale), Stefan Wager (Stanford)
Feedback and Suggestions
If you have feedback or suggestions, please e-mail us at onlinecausalinferenceseminar@gmail.com.
Acknowledgements
We gratefully acknowledge support by the Stanford Department of Statistics and the Stanford Data Science Initiative.
Instructions for Attendees
You can join the webinar by clicking the link on the webpage. If you signed up to the mailing list, you will receive an email with the link before the webinar begins. On Tuesday, you should join the seminar shortly before the start time 8:30 am PT.
Please participate during the seminar!
Due to high demand, we will host the seminar as a Zoom webinar. As an attendee, you will not be able to unmute yourself. If you have questions about the content of the talk, please submit the questions using the Zoom Q&A feature. Time permitting, and depending on the volume of questions, the moderator will either ask your question for you or confirm with you to ask the question yourself and unmute you at a suitable time. In some meetings, the collaborators of the speaker will be online to address your questions in Q&A. Note that Q&A will be moderated by us so you will only be able to see some of the questions of the other attendees. If you want to send messages to the moderators during the seminar, please use the Zoom chat feature.
Zoom instructions
If you have not used Zoom before, we highly recommend downloading and installing the Zoom client before the meeting. Additional instructions on how to use Zoom during a webinar can be found here. Note that for the online causal inference seminar, we do not require registration in advance so you will be able to join by simply clicking the link on this webpage or in the email.
If you have further questions, please drop us an email at onlinecausalinferenceseminar@gmail.com