Ruoxuan Xiong is an assistant professor in the Department of Quantitative Theory and Methods at Emory University. She completed her Ph.D. in Management Science and Engineering from Stanford University in 2020, and was a postdoctoral fellow at the Stanford Graduate School of Business from 2020 to 2021. Her research is at the intersection of econometrics and operations research, focusing on factor modeling, causal inference, and experimental design, and with applications in finance and healthcare. Her work was awarded the Honorable Mention in the 2019 INFORMS George Nicholson Student Paper Competition, and was in the finalists of the 2020 MSOM Student Paper Competition.
| Name of speaker | Ruoxuan Xiong |
| Schedule | 1 April 2022, 10am – 11.30am |
| Link to register | https://nus-sg.zoom.us/meeting/register/tZUpc-uqqzkvHNTjASt7gjozPEloElm7Uuzg |
| Title of talk | Optimal experimental design for staggered rollouts |
| Abstract | In this paper, we study the problem of designing experiments that are conducted on a set of units, such as users or groups of users in an online marketplace, for multiple time periods such as weeks or months. These experiments are particularly useful to study the treatments that have causal effects on both current and future outcomes (instantaneous and lagged effects). The design problem involves selecting a treatment time for each unit, before or during the experiment, in order to most precisely estimate the instantaneous and lagged effects, post experimentation. This optimization of the treatment decisions can directly minimize the opportunity cost of the experiment by reducing its sample size requirement. The optimization is an NP-hard integer program for which we provide a near-optimal solution, when the design decisions are performed all at the beginning (fixed-sample-size designs). Next, we study sequential experiments that allow adaptive decisions during the experiments, and also potentially early stop the experiments, further reducing their cost. However, the sequential nature of these experiments complicates both the design phase and the estimation phase. We propose a new algorithm, PGAE, that addresses these challenges by adaptively making treatment decisions, estimating the treatment effects, and drawing valid post-experimentation inference. PGAE combines ideas from Bayesian statistics, dynamic programming, and sample splitting. Using synthetic experiments on real data sets from multiple domains, we demonstrate that our proposed solutios for fixed-sample-size and sequential experiments reduce the opportunity cost of the experiments by over 50% and 70%, respectively, compared to benchmarks. |
