Robust estimation of optimal dynamic treatment regimes for sequential treatment decisions.

Robust estimation of optimal dynamic treatment regimes for sequential treatment decisions.
复制标题

DOI:
10.1093/biomet/ast014
复制
发表时间:
2013
期刊:
影响因子:
2.7
通讯作者:
Davidian M
Davidian M
中科院分区:
数学2区
文献类型:
--
作者:
Zhang B;Tsiatis AA;Laber EB;Davidian M

文献摘要

参考文献

被引文献

相似文献

动态治疗方案是用于基于患者的病史分配治疗的顺序决策规则的列表。Q-学习和A-学习是用于估计最优机制的两种主要方法,即,使用临床试验或观察性研究的数据,在患者人群中产生最有益的结果。Q-学习需要假设的结果回归模型,而A-学习涉及代表治疗对比和治疗分配的结果回归部分的模型。我们提出了一种替代Q-和A-学习,最大限度地提高了一个双重强大的增广逆概率加权估计的人口平均结果在一个有限的一类制度。仿真结果表明,该方法的性能和鲁棒性的模型误设定,这是一个关键问题。
A dynamic treatment regime is a list of sequential decision rules for assigning treatment based on a patient’s history. Q- and A-learning are two main approaches for estimating the optimal regime, i.e., that yielding the most beneficial outcome in the patient population, using data from a clinical trial or observational study. Q-learning requires postulated regression models for the outcome, while A-learning involves models for that part of the outcome regression representing treatment contrasts and for treatment assignment. We propose an alternative to Q- and A-learning that maximizes a doubly robust augmented inverse probability weighted estimator for population mean outcome over a restricted class of regimes. Simulations demonstrate the method’s performance and robustness to model misspecification, which is a key concern.
DOI: 10.1111/j.1541-0420.2012.01763.x
发表时间: 2012-12
期刊: Biometrics
影响因子: 1.9
作者:
Zhang B;Tsiatis AA;Laber EB;Davidian M
通讯作者: Davidian M
DOI: 10.1198/000313002753631330
发表时间: 2002-02-01
影响因子: 1.8
作者:
Stefanski, LA;Boos, DD
通讯作者: Boos, DD
DOI: 10.1002/sim.2694
发表时间: 2006-12-30
影响因子: 2
作者:
Rosthoj, Susanne;Fullwood, Catherine;Stewart, Syd
通讯作者: Stewart, Syd
DOI: 10.1111/j.1541-0420.2006.00686.x
发表时间: 2007-06-01
期刊: BIOMETRICS
影响因子: 1.9
作者:
Moodie, Erica E. M.;Richardson, Thomas S.;Stephens, David A.
通讯作者: Stephens, David A.
DOI: 10.1214/aos/1176344064
发表时间: 1978-01-01
影响因子: 4.5
作者:
RUBIN, DB
通讯作者: RUBIN, DB