Policy Targeting under Network Interference

This paper discusses the problem of estimating individualized treatment allocation rules under network interference. We propose a method with several appealing features for applications: we let treatment and spillover effects be heterogeneous in the population, and we construct targeting rules that exploit such heterogeneity; we accommodate for arbitrary, possibly non-linear, regression models, and we propose estimators that are robust to model mispecification; treatment allocation rules depend on an arbitrary set of individual, neighbors' and network characteristics, and we allow for general constraints on the policy function and capacity constraints on the number of treated units; the proposed methodology is valid also when only local information of the network is observed. From a theoretical perspective, we establish the first set of guarantees on the utilitarian regret under interference, and we show that it achieves the min-max optimal rate in scenarios of practical and theoretical interest. We provide an exact linear mixed-integer program formulation to the optimization problem, which can be solved using standard optimization routines. We discuss the empirical performance in simulations, and we illustrate our method by investigating the role of social networks in micro-finance decisions.