Events

Prof. Tengyuan Liang
Professor of Econometrics and Statistics
The University of Chicago

Neyman’s seminal paper in 1923, which introduced the potential outcome framework and the analysis of randomized experiments, has arguably laid the foundation of causal inference for cross-sectional data. Consequently, 21st-century medicine embraces a statistical view of effectiveness, considering the implications of treatments and diseases as best understood cross sections. But such population conclusions tell us little about what to do with any particular person. For an individual from whom we collect time-series data, the framework of randomization inference is far less well-understood due to the interference: the potential outcomes at a particular time typically depend on treatments assigned before that time. Motivated by the literature of N-of-1 trials in clinical research and sequential A/B testing in online marketing, in this talk, we study randomization experiments and causal inference for an individual, borrowing insights from system identification and probability theory.

The talk is based on joint work with Benjamin Recht (UC Berkeley).