Who should get vaccinated? Individualized allocation of vaccines over SIR network
利用社交网络数据，在有限疫苗供应下估计个性化分配政策，通过最大化包含溢出效应的公共健康标准，并给出一个计算高效的贪心算法。
How to allocate vaccines over heterogeneous individuals is one of the\nimportant policy decisions in pandemic times. This paper develops a procedure\nto estimate an individualized vaccine allocation policy under limited supply,\nexploiting social network data containing individual demographic\ncharacteristics and health status. We model spillover effects of the vaccines\nbased on a Heterogeneous-Interacted-SIR network model and estimate an\nindividualized vaccine allocation policy by maximizing an estimated social\nwelfare (public health) criterion incorporating the spillovers. While this\noptimization problem is generally an NP-hard integer optimization problem, we\nshow that the SIR structure leads to a submodular objective function, and\nprovide a computationally attractive greedy algorithm for approximating a\nsolution that has theoretical performance guarantee. Moreover, we characterise\na finite sample welfare regret bound and examine how its uniform convergence\nrate depends on the complexity and riskiness of social network. In the\nsimulation, we illustrate the importance of considering spillovers by comparing\nour method with targeting without network information.\n