[论文解读] Efficient Learning of Optimal Individualized Treatment Rules for Heteroscedastic or Misspecified Treatment-Free Effect Models

Weibin Mo, Yufeng Liu|arXiv (Cornell University)|Sep 6, 2021

Advanced Causal Inference Techniques参考文献 86被引用 7

一句话总结

本文提出E-Learning，一种新颖的框架，用于在存在异方差性或治疗效应无处理时的模型误设情况下，估计多臂治疗设置中的最优个体化治疗规则（ITR）。通过将残差方差建模为协变量和治疗的函数，E-Learning实现了半参数效率和双重稳健性，在模型误设或异方差性存在时，显著提升了现有方法的估计效率。

ABSTRACT

Recent development in data-driven decision science has seen great advances in individualized decision making. Given data with individual covariates, treatment assignments and outcomes, researchers can search for the optimal individualized treatment rule (ITR) that maximizes the expected outcome. Existing methods typically require initial estimation of some nuisance models. The double robustness property that can protect from misspecification of either the treatment-free effect or the propensity score has been widely advocated. However, when model misspecification exists, a doubly robust estimate can be consistent but may suffer from downgraded efficiency. Other than potential misspecified nuisance models, most existing methods do not account for the potential problem when the variance of outcome is heterogeneous among covariates and treatment. We observe that such heteroscedasticity can greatly affect the estimation efficiency of the optimal ITR. In this paper, we demonstrate that the consequences of misspecified treatment-free effect and heteroscedasticity can be unified as a covariate-treatment dependent variance of residuals. To improve efficiency of the estimated ITR, we propose an Efficient Learning (E-Learning) framework for finding an optimal ITR in the multi-armed treatment setting. We show that the proposed E-Learning is optimal among a regular class of semiparametric estimates that can allow treatment-free effect misspecification. In our simulation study, E-Learning demonstrates its effectiveness if one of or both misspecified treatment-free effect and heteroscedasticity exist. Our analysis of a Type 2 Diabetes Mellitus (T2DM) observational study also suggests the improved efficiency of E-Learning.

研究动机与目标

• 解决现有ITR估计方法在治疗效应无处理时被误设或结果方差存在异方差性时效率低下的问题。
• 将治疗效应无处理误设和异方差性的双重影响统一为单一来源：协变量和治疗相关的残差方差。
• 开发一种半参数高效估计框架，即使在模型误设下也能保持一致性和最优性。
• 在模拟研究和真实世界观察数据中，展示更高的估计效率和稳健性。

提出的方法

• 将残差方差建模为协变量和治疗的函数，以捕捉异方差性和误设效应。
• 制定加权估计方程，通过逆方差加权来考虑异方差性。
• 通过平衡权重整合倾向得分信息，实现对治疗效应无处理模型和倾向得分模型的双重稳健性。
• 采用两阶段估计程序：首先估计方差函数，然后应用加权最小二乘法估计最优ITR。
• 确保估计量在允许治疗效应无处理误设的常规估计类中具有半参数效率。
• 利用增广逆概率加权估计方程（AIPWE）框架，以增强稳健性和效率。

实验结果

研究问题

• RQ1 结果残差中的异方差性如何影响最优ITR估计的效率？
• RQ2 治疗效应无处理误设和异方差性能否在单一统计框架下正式统一？
• RQ3 一种同时考虑异方差性的双重稳健ITR估计量是否比现有方法具有更高的效率？
• RQ4 当干扰模型正确设定时，所提出的E-Learning框架是否具有半参数效率？
• RQ5 当治疗效应无处理模型和倾向得分模型均被误设时，E-Learning在有限样本中的表现如何？

主要发现

• 当治疗效应无处理模型和倾向得分模型均正确设定时，E-Learning实现了半参数效率。
• 在模拟研究中，无论在同方差还是异方差误差结构下，E-Learning在误分类率和遗憾值方面均显著低于Q-Learning、D-Learning和RD-Learning。
• 在模型误设情况下，E-Learning保持了更优性能，在高维设置中误分类率比竞争方法低20%至40%。
• 在T2DM观察性研究中，E-Learning表现出更高的估计稳定性与更低的遗憾值，系数估计在不同数据配置下更一致。
• 该框架的双重稳健性特性确保了即使其中一个干扰模型（治疗效应无处理或倾向得分）被误设，估计仍保持一致。
• 对ACTG175数据集的残差分析证实，同时存在治疗效应无处理误设和异方差性，验证了所提出框架的必要性。