A0677
Title: Causal clustering: Design of cluster experiments under network interference
Authors: Lihua Lei - Stanford University (United States) [presenting]
Abstract: The design of cluster experiments is studied to estimate the global treatment effect in the presence of network spillovers. A framework is provided to choose the clustering that minimizes the worst-case mean-squared error of the estimated global effect. It is shown that optimal clustering solves a novel penalized min-cut optimization problem computed via off-the-shelf semi-definite programming algorithms. The analysis also characterizes simple conditions to choose between any two cluster designs, including choosing between a cluster or individual-level randomization. The method's properties are illustrated using unique network data from the universe of Facebook's users and existing data from a field experiment.
