{"title":"带隐变量的因果模型辨识。","authors":"Ilya Shpitser","doi":"","DOIUrl":null,"url":null,"abstract":"<p><p>Targets of inference that establish causality are phrased in terms of counterfactual responses to interventions. These <i>potential outcomes</i> operationalize cause effect relationships by means of comparisons of cases and controls in hypothetical randomized controlled experiments. In many applied settings, data on such experiments is not directly available, necessitating assumptions linking the counterfactual target of inference with the factual observed data distribution. This link is provided by causal models. Originally defined on potential outcomes directly (Rubin, 1976), causal models have been extended to longitudinal settings (Robins, 1986), and reformulated as graphical models (Spirtes et al., 2001; Pearl, 2009). In settings where common causes of all observed variables are themselves observed, many causal inference targets are identified via variations of the expression referred to in the literature as the <i>g-formula</i> (Robins, 1986), the <i>manipulated distribution</i> (Spirtes et al., 2001), or the <i>truncated factorization</i> (Pearl, 2009). In settings where hidden variables are present, identification results become considerably more complicated. In this manuscript, we review identification theory in causal models with hidden variables for common targets that arise in causal inference applications, including causal effects, direct, indirect, and path-specific effects, and outcomes of dynamic treatment regimes. We will describe a simple formulation of this theory (Tian and Pearl, 2002; Shpitser and Pearl, 2006b,a; Tian, 2008; Shpitser, 2013) in terms of causal graphical models, and the fixing operator, a statistical analogue of the intervention operation (Richardson et al., 2017).</p>","PeriodicalId":44492,"journal":{"name":"Journal of the SFdS","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7685307/pdf/nihms-1063757.pdf","citationCount":"0","resultStr":"{\"title\":\"Identification in Causal Models With Hidden Variables.\",\"authors\":\"Ilya Shpitser\",\"doi\":\"\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Targets of inference that establish causality are phrased in terms of counterfactual responses to interventions. These <i>potential outcomes</i> operationalize cause effect relationships by means of comparisons of cases and controls in hypothetical randomized controlled experiments. In many applied settings, data on such experiments is not directly available, necessitating assumptions linking the counterfactual target of inference with the factual observed data distribution. This link is provided by causal models. Originally defined on potential outcomes directly (Rubin, 1976), causal models have been extended to longitudinal settings (Robins, 1986), and reformulated as graphical models (Spirtes et al., 2001; Pearl, 2009). In settings where common causes of all observed variables are themselves observed, many causal inference targets are identified via variations of the expression referred to in the literature as the <i>g-formula</i> (Robins, 1986), the <i>manipulated distribution</i> (Spirtes et al., 2001), or the <i>truncated factorization</i> (Pearl, 2009). In settings where hidden variables are present, identification results become considerably more complicated. In this manuscript, we review identification theory in causal models with hidden variables for common targets that arise in causal inference applications, including causal effects, direct, indirect, and path-specific effects, and outcomes of dynamic treatment regimes. We will describe a simple formulation of this theory (Tian and Pearl, 2002; Shpitser and Pearl, 2006b,a; Tian, 2008; Shpitser, 2013) in terms of causal graphical models, and the fixing operator, a statistical analogue of the intervention operation (Richardson et al., 2017).</p>\",\"PeriodicalId\":44492,\"journal\":{\"name\":\"Journal of the SFdS\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7685307/pdf/nihms-1063757.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of the SFdS\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2020/6/30 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of the SFdS","FirstCategoryId":"1085","ListUrlMain":"","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2020/6/30 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
建立因果关系的推理目标是根据对干预的反事实反应来表述的。这些潜在结果通过在假设的随机对照实验中对病例和对照进行比较来实现因果关系的操作性。在许多应用环境中,此类实验的数据不能直接获得,因此需要将反事实的推理目标与事实观察到的数据分布联系起来。这种联系是由因果模型提供的。最初直接定义潜在结果(Rubin, 1976),因果模型已经扩展到纵向设置(Robins, 1986),并重新制定为图形模型(Spirtes等人,2001;珍珠,2009)。在观察到所有被观察变量的共同原因的情况下,通过文献中提到的g公式(Robins, 1986)、操纵分布(Spirtes et al., 2001)或截断因子分解(Pearl, 2009)等表达式的变化来确定许多因果推理目标。在存在隐藏变量的设置中,识别结果变得相当复杂。在本文中,我们回顾了在因果推理应用中出现的具有隐藏变量的因果模型中的识别理论,包括因果效应、直接效应、间接效应和路径特异性效应,以及动态治疗方案的结果。我们将描述这一理论的一个简单公式(Tian and Pearl, 2002;Shpitser and Pearl, 2006b,a;田,2008;Shpitser, 2013)在因果图模型方面,以及固定算子,干预操作的统计模拟(Richardson等,2017)。
Identification in Causal Models With Hidden Variables.
Targets of inference that establish causality are phrased in terms of counterfactual responses to interventions. These potential outcomes operationalize cause effect relationships by means of comparisons of cases and controls in hypothetical randomized controlled experiments. In many applied settings, data on such experiments is not directly available, necessitating assumptions linking the counterfactual target of inference with the factual observed data distribution. This link is provided by causal models. Originally defined on potential outcomes directly (Rubin, 1976), causal models have been extended to longitudinal settings (Robins, 1986), and reformulated as graphical models (Spirtes et al., 2001; Pearl, 2009). In settings where common causes of all observed variables are themselves observed, many causal inference targets are identified via variations of the expression referred to in the literature as the g-formula (Robins, 1986), the manipulated distribution (Spirtes et al., 2001), or the truncated factorization (Pearl, 2009). In settings where hidden variables are present, identification results become considerably more complicated. In this manuscript, we review identification theory in causal models with hidden variables for common targets that arise in causal inference applications, including causal effects, direct, indirect, and path-specific effects, and outcomes of dynamic treatment regimes. We will describe a simple formulation of this theory (Tian and Pearl, 2002; Shpitser and Pearl, 2006b,a; Tian, 2008; Shpitser, 2013) in terms of causal graphical models, and the fixing operator, a statistical analogue of the intervention operation (Richardson et al., 2017).