Josefsson Maria, Daniels Michael J
Centre for Demographic and Ageing Research, Umeå University, Sweden.
Department of Statistics, University of Florida.
J R Stat Soc Ser C Appl Stat. 2021 Mar;70(2):398-414. doi: 10.1111/rssc.12464. Epub 2021 Jan 6.
Causal inference with observational longitudinal data and time-varying exposures is often complicated by time-dependent confounding and attrition. The G-computation formula is one approach for estimating a causal effect in this setting. The parametric modeling approach typically used in practice relies on strong modeling assumptions for valid inference, and moreover depends on an assumption of missing at random, which is not appropriate when the missingness is missing not at random (MNAR) or due to death. In this work we develop a flexible Bayesian semi-parametric G-computation approach for assessing the causal effect on the subpopulation that would survive irrespective of exposure, in a setting with MNAR dropout. The approach is to specify models for the observed data using Bayesian additive regression trees, and then use assumptions with embedded sensitivity parameters to identify and estimate the causal effect. The proposed approach is motivated by a longitudinal cohort study on cognition, health, and aging, and we apply our approach to study the effect of becoming a widow on memory. We also compare our approach to several standard methods.
利用观察性纵向数据和随时间变化的暴露因素进行因果推断，常常会因随时间变化的混杂因素和失访而变得复杂。G计算法则是在此种情况下估计因果效应的一种方法。实践中通常使用的参数建模方法依赖于强建模假设以进行有效推断，而且还依赖于随机缺失假设，当缺失不是随机的（MNAR）或由于死亡导致缺失时，该假设并不适用。在这项工作中，我们开发了一种灵活的贝叶斯半参数G计算方法，用于在存在MNAR失访的情况下，评估对无论暴露情况如何都能存活的亚人群的因果效应。该方法是使用贝叶斯加法回归树为观察到的数据指定模型，然后使用带有嵌入式敏感性参数的假设来识别和估计因果效应。所提出的方法是受一项关于认知、健康和衰老的纵向队列研究的启发，并且我们将我们的方法应用于研究成为寡妇对记忆的影响。我们还将我们的方法与几种标准方法进行了比较。
