scholarly journals Causal inference under multiple versions of treatment

2013 ◽  
Vol 1 (1) ◽  
pp. 1-20 ◽  
Author(s):  
Tyler J. VanderWeele ◽  
Miguel A. Hernan
Keyword(s):  
Causal Inference ◽  
Treatment Effect ◽  
Indirect Effects ◽  
Causal Effect ◽  
Potential Outcomes ◽  
Direct And Indirect Effects ◽  
Treatment Variable ◽  
Effect Of Treatment ◽  
New Treatment ◽  
Potential Outcomes Framework

Abstract: In this article, we discuss causal inference when there are multiple versions of treatment. The potential outcomes framework, as articulated by Rubin, makes an assumption of no multiple versions of treatment, and here we discuss an extension of this potential outcomes framework to accommodate causal inference under violations of this assumption. A variety of examples are discussed in which the assumption may be violated. Identification results are provided for the overall treatment effect and the effect of treatment on the treated when multiple versions of treatment are present and also for the causal effect comparing a version of one treatment to some other version of the same or a different treatment. Further identification and interpretative results are given for cases in which the version precedes the treatment as when an underlying treatment variable is coarsened or dichotomized to create a new treatment variable for which there are effectively “multiple versions”. Results are also given for effects defined by setting the version of treatment to a prespecified distribution. Some of the identification results bear resemblance to identification results in the literature on direct and indirect effects. We describe some settings in which ignoring multiple versions of treatment, even when present, will not lead to incorrect inferences.

scholarly journals Emulating Target Trials to Improve Causal Inference from Agent-Based Models

2021 ◽  
Author(s):  
Eleanor J Murray ◽  
Brandon D L Marshall ◽  
Ashley L Buchanan
Keyword(s):  
Causal Inference ◽  
Disease Transmission ◽  
Causal Effect ◽  
Potential Outcomes ◽  
Emergent Properties ◽  
Target Trial ◽  
Agent Based Models ◽  
Agent Based ◽  
Effect Estimation ◽  
Potential Outcomes Framework

Abstract Agent-based models are a key tool for investigating the emergent properties of population health settings, such as infectious disease transmission, where the exposure often violates the key ‘no interference’ assumption of traditional causal inference under the potential outcomes framework. Agent-based models and other simulation-based modeling approaches have generally been viewed as a separate knowledge-generating paradigm from the potential outcomes framework, but this can lead to confusion about how to interpret the results of these models in real-world settings. By explicitly incorporating the target trial framework into the development of an agent-based or other simulation model, we can clarify the causal parameters of interest, as well as make explicit the assumptions required for valid causal effect estimation within or between populations. In this paper, we describe the use of the target trial framework for designing agent-based models when the goal is estimation of causal effects in the presence of interference, or spillover.

scholarly journals Gformula: Estimating Causal Effects in the Presence of Time-Varying Confounding or Mediation using the G-Computation Formula

2011 ◽  
Vol 11 (4) ◽  
pp. 479-517 ◽  
Author(s):  
Rhian M. Daniel ◽  
Bianca L. De Stavola ◽  
Simon N. Cousens
Keyword(s):  
Indirect Effects ◽  
Related Problem ◽  
Causal Effect ◽  
Causal Effects ◽  
Time Varying ◽  
Direct And Indirect Effects ◽  
Computation Procedure ◽  
Computation Formula ◽  
Intermediate Variables

This article describes a new command, gformula, that is an implementation of the g-computation procedure. It is used to estimate the causal effect of time-varying exposures on an outcome in the presence of time-varying confounders that are themselves also affected by the exposures. The procedure also addresses the related problem of estimating direct and indirect effects when the causal effect of the exposures on an outcome is mediated by intermediate variables, and in particular when confounders of the mediator–outcome relationships are themselves affected by the exposures. A brief overview of the theory and a description of the command and its options are given, and illustrations using two simulated examples are provided.

scholarly journals Invited Commentary: Counterfactuals in Social Epidemiology—Thinking Outside of “the Box”

2019 ◽  
Vol 189 (3) ◽  
pp. 175-178 ◽  
Author(s):  
Tyler J VanderWeele
Keyword(s):  
Social Movements ◽  
Causal Inference ◽  
Social Determinants Of Health ◽  
Social Determinants ◽  
Social Epidemiology ◽  
Determinants Of Health ◽  
Potential Outcomes ◽  
The Social ◽  
Potential Outcomes Framework

Abstract There are tensions inherent between many of the social exposures examined within social epidemiology and the assumptions embedded in quantitative potential-outcomes-based causal inference framework. The potential-outcomes framework characteristically requires a well-defined hypothetical intervention. As noted by Galea and Hernán (Am J Epidemiol. 2020;189(3):167–170), for many social exposures, such well-defined hypothetical exposures do not exist or there is no consensus on what they might be. Nevertheless, the quantitative potential-outcomes framework can still be useful for the study of some of these social exposures by creative adaptations that 1) redefine the exposure, 2) separate the exposure from the hypothetical intervention, or 3) allow for a distribution of hypothetical interventions. These various approaches and adaptations are reviewed and discussed. However, even these approaches have their limits. For certain important historical and social determinants of health such as social movements or wars, the quantitative potential-outcomes framework with well-defined hypothetical interventions is the wrong tool. Other modes of inquiry are needed.

scholarly journals Treatment Effects on Ordinal Outcomes: Causal Estimands and Sharp Bounds

2018 ◽  
Vol 43 (5) ◽  
pp. 540-567 ◽  
Author(s):  
Jiannan Lu ◽  
Peng Ding ◽  
Tirthankar Dasgupta
Keyword(s):  
Causal Effect ◽  
Real Life ◽  
Causal Effects ◽  
Potential Outcomes ◽  
Randomized Experiments ◽  
Sharp Bounds ◽  
Average Causal Effect ◽  
Behavioral Studies ◽  
And Control ◽  
Potential Outcomes Framework

Assessing the causal effects of interventions on ordinal outcomes is an important objective of many educational and behavioral studies. Under the potential outcomes framework, we can define causal effects as comparisons between the potential outcomes under treatment and control. However, unfortunately, the average causal effect, often the parameter of interest, is difficult to interpret for ordinal outcomes. To address this challenge, we propose to use two causal parameters, which are defined as the probabilities that the treatment is beneficial and strictly beneficial for the experimental units. However, although well-defined for any outcomes and of particular interest for ordinal outcomes, the two aforementioned parameters depend on the association between the potential outcomes and are therefore not identifiable from the observed data without additional assumptions. Echoing recent advances in the econometrics and biostatistics literature, we present the sharp bounds of the aforementioned causal parameters for ordinal outcomes, under fixed marginal distributions of the potential outcomes. Because the causal estimands and their corresponding sharp bounds are based on the potential outcomes themselves, the proposed framework can be flexibly incorporated into any chosen models of the potential outcomes and is directly applicable to randomized experiments, unconfounded observational studies, and randomized experiments with noncompliance. We illustrate our methodology via numerical examples and three real-life applications related to educational and behavioral research.

scholarly journals Instrumental variables based on twin births are by definition not valid (v.3.0)

2017 ◽  
Author(s):  
Stefan Öberg
Keyword(s):  
Instrumental Variables ◽  
Natural Experiment ◽  
Causal Effect ◽  
Potential Outcomes ◽  
Number Of Children ◽  
Biased Estimates ◽  
Potential Outcomes Framework

Twin births are a well-known and widespread example of a so-called “natural experiment”. Instrumental variables based on twin births have been used in many studies to estimate the causal effect of the number of children on the parents or siblings. I use the potential outcomes framework to show that these instrumental variables do not work as intended. They are fundamentally flawed and will always lead to severely biased estimates without any meaningful interpretation. This has been overlooked in previous research because too little attention has been paid to defining the treatment in this natural experiment. I analyze three different possible interpretations of the treatment and show that they all lead to inherent violations of the necessary assumptions. The effect of the number of on the parents or siblings is a policy relevant and theoretically important issue. The scientific record should therefore be corrected to not lead to misguided decisions.

Causal Inference with Networked Treatment Diffusion

2018 ◽  
Vol 48 (1) ◽  
pp. 152-181 ◽  
Author(s):  
Weihua An
Keyword(s):  
Causal Inference ◽  
Treatment Effect ◽  
Simple Structure ◽  
Smoking Prevention ◽  
Causal Effects ◽  
Potential Outcomes ◽  
Statistical Inferences ◽  
Social Settings ◽  
Biased Estimates

Treatment interference (i.e., one unit’s potential outcomes depend on other units’ treatment) is prevalent in social settings. Ignoring treatment interference can lead to biased estimates of treatment effects and incorrect statistical inferences. Some recent studies have started to incorporate treatment interference into causal inference. But treatment interference is often assumed to follow a simple structure (e.g., treatment interference exists only within groups) or measured in a simplistic way (e.g., only based on the number of treated friends). In this paper, I highlight the importance of collecting data on actual treatment diffusion in order to more accurately measure treatment interference. Furthermore, I show that with accurate measures of treatment interference, we can identify and estimate a series of causal effects that are previously unavailable, including the direct treatment effect, treatment interference effect, and treatment effect on interference. I illustrate the methods through a case study of a social network–based smoking prevention intervention.

Causal Inference

Sociology ◽  
2020 ◽  
Author(s):  
Pablo Geraldo Bastías ◽  
Jennie E. Brand
Keyword(s):  
Causal Inference ◽  
Statistical Methods ◽  
Quantitative Methods ◽  
Fundamental Problem ◽  
Causal Effect ◽  
College Education ◽  
Directed Acyclic Graphs ◽  
Structural Equations ◽  
Potential Outcomes ◽  
Formal Representation

Causal inference is a growing interdisciplinary subfield in statistics, computer science, economics, epidemiology, and the social sciences. In contrast with both traditional quantitative methods and cutting-edge approaches like machine learning, causal inference questions are defined in relation to potential outcomes, or variable values that are counterfactual to the observed world and therefore cannot be answered from joint probabilities alone, even with infinite data. The fact that one can possibly observe at most one potential outcome among those of interest is known as the “fundamental problem of causal inference.” For example, in this framework, the economic return to college education can be defined as a comparison between two potential outcomes: the wages of an individual with a college education versus the wages that the same individual would have received had he or she not attended college. In general, researchers are interested in estimating such effects for certain groups and comparing the effects for different subpopulations. Critical to causal inference is recognizing that, to answer causal questions from observed data, one has to rely on untestable assumptions about how the data were generated. In other words, there is no particular statistical method that would render a conclusion “causal”; the validity of such an interpretation depends on a combination of data, assumptions about the data-generating process based on expert judgment, and estimation techniques. In the last several decades, our understanding of causality has improved enormously, owing to a conceptual apparatus and a mathematical language that enables rigorous conceptualization of causal quantities and formal representation of causal assumptions, while still employing familiar statistical methods. Potential outcomes or the Neyman-Rubin causal model and structural equations encoded as directed acyclic graphs (DAGs, also known as structural causal models) are two common approaches for conceptualizing causal relationships. The symbiosis of both languages offers a powerful framework to address causal questions. This review covers developments in both causal identification (i.e., deciding if a quantity of interest would be recoverable from infinite data, based on our assumptions) and causal effect estimation (i.e., the use of statistical methods to approximate that answer with finite, although potentially big, data). The literature is presented following the type of assumptions and questions frequently encountered in empirical research, ending with a discussion of promising new directions in the field.

Beyond LATE: Estimation of the Average Treatment Effect with an Instrumental Variable

2013 ◽  
Vol 21 (4) ◽  
pp. 492-506 ◽  
Author(s):  
Peter M. Aronow ◽  
Allison Carnegie
Keyword(s):  
Treatment Effect ◽  
Instrumental Variable ◽  
Causal Effect ◽  
Average Treatment Effect ◽  
Treatment Variable ◽  
Estimation Strategy ◽  
Average Treatment ◽  
Local Average Treatment Effect ◽  
Receive Treatment ◽  
Local Average

Political scientists frequently use instrumental variables (IV) estimation to estimate the causal effect of an endogenous treatment variable. However, when the treatment effect is heterogeneous, this estimation strategy only recovers the local average treatment effect (LATE). The LATE is an average treatment effect (ATE) for a subset of the population: units that receive treatment if and only if they are induced by an exogenous IV. However, researchers may instead be interested in the ATE for the entire population of interest. In this article, we develop a simple reweighting method for estimating the ATE, shedding light on the identification challenge posed in moving from the LATE to the ATE. We apply our method to two published experiments in political science in which we demonstrate that the LATE has the potential to substantively differ from the ATE.

scholarly journals Deep Learning of Potential Outcomes

2021 ◽  
Author(s):  
Bernard Koch ◽  
Tim Sainburg ◽  
Pablo Geraldo ◽  
SONG JIANG ◽  
Yizhou Sun ◽  
...  
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Causal Inference ◽  
Deep Neural Networks ◽  
Linear Time ◽  
Potential Outcomes ◽  
Time Varying ◽  
Heterogeneous Treatment Effects ◽  
Sharp Focus ◽  
Potential Outcomes Framework

This review systematizes the emerging literature for causal inference using deep neural networks under the potential outcomes framework. It provides an intuitive introduction on how deep learning can be used to estimate/predict heterogeneous treatment effects and extend causal inference to settings where confounding is non-linear, time varying, or encoded in text, networks, and images. To maximize accessibility, we also introduce prerequisite concepts from causal inference and deep learning. The survey differs from other treatments of deep learning and causal inference in its sharp focus on observational causal estimation, its extended exposition of key algorithms, and its detailed tutorials for implementing, training, and selecting among deep estimators in Tensorflow 2 available at github.com/kochbj/Deep-Learning-for-Causal-Inference.

Sign in / Sign up

Export Citation Format

Share Document