Panagiotis (Panos) Toulis studies causal inference and experimental design in such complex systems as multi-agent economies and social networks, in order to evaluate the efficacy of interventions on such systems. Applications of this research include market design and policy analysis. He focuses on three distinct problems—interference, entanglement, and dynamics—each of which can invalidate classical causal methods. He is also interested in the interface of statistics and optimization, particularly the principled statistical analysis of large datasets, with a focus on implicit stochastic approximation methods, which are numerically stable.
Toulis’s research has been published in the Annals of Statistics, Games of Economic Behavior, and Statistics and Computing, and in conjunction with major machine learning and economics conferences. He received the Arthur P. Dempster Award from Harvard University’s Department of Statistics, which is given annually to a PhD student who has made significant contributions to theoretical or foundational research in statistics. Other honors include a LinkedIn Economic Graph Challenge award and a Google United States/Canada PhD Fellowship in statistics.
Toulis holds a BS in electrical and computer engineering from Aristotle University in Thessaloniki, Greece, as well as MS degrees in statistics and computer science and a PhD in statistics from Harvard University. He also has prior corporate experience in software engineering at Google Inc. and at several start-up companies in Greece.