报告主题：Evaluation of dynamic treatment regimes with infinite and irregular observations

时间：2023年4月14日13:30-14:30

地点：上川路校区五教117/腾讯会议  会议号：984 1292 8754 密码：6666

报告人：陈欣

Dynamic treatment regimes are sequential decision rules assigning individualized treatments to patients based on evolving treatments and covariate history. This talk focuses on evaluating dynamic treatment regimes with infinite and irregular observations. Following the reinforcement learning techniques, a state-action value function is proposed to measure the goodness of a policy, and Bellman equations are constructed to estimate the proposed value function. Simulation studies and a real data application are conducted to illustrate the proposed methods.