Q-learning is a regression-based approach that is widely used to formalize the development of an optimal dynamic treatment strategy. Finite dimensional working models are typically used to estimate certain nuisance parameters, and misspecification of these working models can result in residual confounding and/or efficiency loss. We propose a robust Q-learning approach which allows estimating such nuisance parameters using data-adaptive techniques. We study the asymptotic behavior of our estimators and provide simulation studies that highlight the need for and usefulness of the proposed method in practice. We use the data from the "Extending Treatment Effectiveness of Naltrexone" multi-stage randomized trial to illustrate our proposed methods.
A unifying approach for doubly-robust $\ell_1$ regularized estimation of causal contrasts
We consider inference about a scalar parameter under a non-parametric model based on a one-step estimator computed as a plug in estimator plus the empirical mean of an estimator of the parameter's influence function . We focus on a class of parameters that have influence function which depends on two infinite dimensional nuisance functions and such that the bias of the one-step estimator of the parameter of interest is the expectation of the product of the estimation errors of the two nuisance functions. Our class includes many important treatment effect contrasts of interest in causal inferen…
A Note on Debiased/Double Machine Learning Logistic Partially Linear Model
Logistic部分线性模型的双鲁棒推理在许多应用领域中具有特殊的意义。最近,Tan(2019)为此提出了一种简单灵活的双稳健估计器。他们引入了两个滋扰模型,即logistic模型中的非参数分量和暴露协变量的条件平均值(给定其他协变量和固定响应),并将其指定为固定维参数模型。他们的框架可能会扩展到机器学习或最近开发的高维滋扰建模,例如,在Zhuzhukovet al.(2018 a,b)和Smucler et al.(2019); Tan(2020)。基于此,本文提出了去偏/双机器学习logistic部分线性模型。对于滋扰模型的构建,我们分别考虑使用高维稀疏参数模型和一般的机器学习方法。通过推导一定的矩方程来校正滋扰模型的一阶偏差,我们保持了高维超稀疏滋扰模型的模型双重鲁棒性。我们还讨论并比较了我们的方法与去偏LASSO(Van deGeer等人,2014年)。为了实现机器学习的建议,我们设计了一个完整的模型改装程序,允许使用任何黑盒条件均值估计方法在我们的框架。在机器学习设置下,我们的方法在类似意义上具有双倍的鲁棒性,如Zerichzhukov et al.(2018 a)。
Why Is the Double-Robust Estimator for Causal Inference Not Doubly Robust for Variance Estimation?
Vanderbilt University Medical Center(范德堡大学医学中心);UCSD Herbert Wertheim School of Public Health and Human Longevity Science(加州大学圣地亚哥分校赫伯特·韦特海姆公共卫生与人类长寿科学学院);University of Florida(佛罗里达大学)
Bootstrap test procedure for variance components in nonlinear mixed effects models in the presence of nuisance parameters and a singular Fisher Information Matrix
Doubly Robust Functional Representation Learning for Longitudinal Causal Inference with Irregular Histories
Academy of Life and Natural Sciences, Xi’an Jiaotong-Liverpool University(西交利物浦大学生命与自然科学学院);Aerospace Information Research Institute, Chinese Academy of Sciences(中国科学院空天信息研究院)
Doubly Robust Goodness-of-Fit Test of Coarse Structural Nested Mean Models with Application to Initiating combination antiretroviral treatment in HIV-Positive Patients
Debiased machine learning for combining probability and non-probability survey data
MRC Biostatistics Unit, University of Cambridge(剑桥大学医学研究委员会生物统计学单元);Department of Genetics, University of Cambridge(剑桥大学生物学系);Institute for Global Health, University College London(伦敦大学学院全球健康研究所)
Dynamic treatment effects: high-dimensional inference under model misspecification
Institute of Statistics and Big Data, Renmin University of China(中国人民大学统计与大数据研究院);School of Statistics and Management, Shanghai University of Finance and Economics(上海财经大学统计与管理学院);University of California, San Diego(加利福尼亚大学圣迭戈分校)
Doubly Robust Proximal Causal Learning for Continuous Treatments
Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University(复旦大学类脑智能科学与技术研究院);School of Data Science, Fudan University(复旦大学数据科学学院);Zhangjiang Fudan International Innovation Center(张江复旦国际创新中心);Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence (Fudan University)(复旦大学计算神经科学与类脑智能重点实验室);MOE Frontiers Center for Brain Science, Fudan University(教育部脑科学前沿中心(复旦大学));Shanghai Engineering Research Center of AI & Robotics, Fudan University(上海市人工智能与机器人工程技术研究中心(复旦大学));Engineering Research Center of AI & Robotics, Ministry of Education, Fudan University(教育部人工智能与机器人工程研究中心(复旦大学))
Collaborative Indirect Treatment Comparisons with Multiple Distributed Single-arm Trials
University of Pennsylvania(宾夕法尼亚大学);Pfizer Inc.(辉瑞公司);Division of Biostatistics and Health Data Science, University of Minnesota Twin Cities(明尼苏达双城大学生物统计与健康数据科学系);School of Mathematical Sciences, Peking University(北京大学数学科学学院)