Presentations

Optimal Pre-Analysis Plans: Statistical Decisions Subject to Implementability
October 2023, Yale (Cowles Foundation)
Slides

Economics and Machine Learning: What can they teach each other?
September 2023, Berlin
Slides
Recording

Employing the unemployed of Marienthal: Evaluation of a guaranteed job program
March 2023, IHS
Slides
Recording

The political economy of AI: Who controls the means of prediction?
February 2023, Oxford series “What economists really do”
Slides
Recording
April 2023, MIT SERC Symposium
Recording

Democratic control of the means of prediction
October 2022, Oxford Ethics in AI Colloquium
Slides
Recording

Adaptive maximization of social welfare in theory and practice
October 2022, Harvard and MIT
Slides

Discussion of:
Increasing the uptake of long-acting reversible contraceptives among adolescents and young women in Cameroon

July 2022, NBER Summer Institute
Slides

Adaptive maximization of social welfare
June 2022, ENSAI Artificial Intelligence and Economic Decision Making conference, Rennes
Slides
Recording at ENSAI 2022

Rationalizing Pre-Analysis Plans: Statistical Decisions Subject to Implementability
January 2022
Slides
Recording for BITSS 2022

Employing the unemployed of Marienthal:
Evaluation of a guaranteed job program

January 2022, Marie Jahoda memorial conference
Slides

How to run an adaptive field experiment
December 2021, Field Days conference
Slides

Machine learning for policy
November 2021, deputy ministers of Pakistan
Slides

Learning by matching
September 2021, EAAMO conference
Slides

Social foundations for statistics and machine learning
London, September 2021
Slides

Piloting new social safety nets: Evaluation of a job guarantee program and of a basic income program
Vienna, June 2021
Slides
Recording

Statistical decision theory cannot justify
randomization or pre-registration for experiments.

Sorbonne, Philosophy of Science Workshop, June 2021
Slides

Philosophical questions about machine learning theory:
Online Learning, Bandit Algorithms, and Reinforcement Learning

Harvard Philosophy of Causality Workshop, March 2021
Slides

Fairness, equality, and power in algorithmic decision making
Oxford Law School, March 2021
Slides
Recording for Informs 2021

Probably approximately correct learning and adversarial online learning
Alan Turing Institute, Economics and ML reading group, February 2021
Slides I
Slides II

The social impact of algorithmic decision making: Economic perspectives
Stanford, October 2020
Slides

Basic income and basic income experiments
July 2020
Slides
Recording

What do we want? And when do we want it?
Alternative objectives and their implications for experimental design.

May 2020
Slides
Recording of presentation in the Chamberlain online seminar
(With discussions by David McKenzie and Max Tabord-Meehan)

Adaptive combinatorial allocation: How to use limited resources while learning what works
May 2020
Slides

Fairness, equality, and power in algorithmic decision making
May 2020
Slides

No data in the void
Economics for Inclusive Prosperity Conference, Harvard Kennedy School, March 27, 2020
Slides

Adaptive treatment assignment in experiments for policy choice
Causal Machine Learning Workshop St. Gallen, January 20, 2020
Slides

Experiments designed to help the participants
IRC Jordan, July 10, 2019
Slides

Conference “Statistics in a social context”
Harvard, May 10-11, 2019
Conference announcement, schedule, and links to talks

Adaptive treatment assignment in experiments for policy choice (short version)
MIT, May 18, 2019
Slides

Statistics in a Social Context - Opening remarks
Harvard, May 10, 2019
Slides

Designing basic income experiments
Jain Family Institute, April 12, 2019
Slides

Machine learning, shrinkage estimation, and economic theory
Sciences Po, Friday, December 14, 2018
Slides

Which findings get published? Which findings should be published? (short version)
BITSS Annual Meeting, UC Berkeley, Monday, December 10, 2018
Slides
Video

Causality and randomization
Harvard Philosophy Causality Conference, Friday, November 2, 2018
Slides

Which findings get published? Which findings should be published?
Bocconi, Tuesday, September 4, 2018
Slides

Which findings should be published
Cologne, Thursday, August 30, 2018
Slides

How to use economic theory to improve estimators
Montreal, Wednesday, June 27, 2018
Slides

Optimal taxation and insurance using machine learning
Machine Learning in Economics and Econometrics conference, Munich, Tuesday, May 29, 2018
Slides

Habilitationsvortrag: Machine learning, shrinkage estimation, and economic theory
WU Wien, Friday, May 25, 2018
Slides

Estimating Risk
Gary Chamberlain Retirement Conference, Friday, May 4, 2018
Slides

“Approximate Cross-Validation,” and “Dynamic Experiments for Policy Choice”
Harvard, Monday, April 23, 2018
Slides

Choosing among regularized estimators in empirical economics
AEA Session: Machine Learning for Policy Research, Friday, January 5, 2018
Slides

Identification of and correction for publication bias
Sciences Po, Wednesday, December 13, 2017
Slides

Technischer Wandel und der Arbeitsmarkt
Hauptverband der österreichischen Sozialversicherungsträger, Friday, October 20, 2017
Slides

Empirical research on economic inequality: Normative considerations and empirical practice
IHS Wien, Monday, May 15, 2017
Slides

International corporate taxation: Problems and a reform proposal
Re:publika Berlin, Monday, May 8, 2017
Slides

Uniformity and the delta-method
Monday, October 27, 2014
Slides

Who wins, who loses? Tools for distributional policy evaluation
Friday, September 26, 2014
Slides
Video

Why experimenters should not randomize, and what they should do instead
Wednesday, February 27, 2013
Slides
Video

Partial identification, distributional preferences, and the welfare ranking of policies
Friday, April 27, 2012
Slides

Nonparametric inference on the number of equilibria
Thursday, May 5, 2011
Slides

Identification in a model of sorting with social externalities and the causes of urban segregation
Tuesday, March 15, 2011
Slides

Identification in Triangular Systems Using Control Functions
Wednesday, November 10, 2010
Slides