scholarly journals Stein-like estimators for causal mediation analysis in randomized trials

2019 ◽  
Vol 29 (4) ◽  
pp. 1129-1148 ◽  
Author(s):  
Cedric E Ginestet ◽  
Richard Emsley ◽  
Sabine Landau
Keyword(s):  
Least Squares ◽  
Mediation Analysis ◽  
Indirect Effects ◽  
Ordinary Least Squares ◽  
Least Squares Estimator ◽  
Two Stage ◽  
Causal Mediation ◽  
Causal Mediation Analysis ◽  
Two Stage Least Squares ◽  
The Impact

Causal mediation analysis aims to estimate natural direct and natural indirect effects under clearly specified assumptions. Traditional mediation analysis based on Ordinary Least Squares assumes an absence of unmeasured causes to the putative mediator and outcome. When these assumptions cannot be justified, instrumental variable estimators can be used in order to produce an asymptotically unbiased estimator of the mediator-outcome link, commonly referred to as a Two-Stage Least Squares estimator. Such bias removal, however, comes at the cost of variance inflation. A Semi-Parametric Stein-Like estimator has been proposed in the literature that strikes a natural trade-off between the unbiasedness of the Two-Stage Least Squares procedure and the relatively small variance of the Ordinary Least Squares estimator. The Semi-Parametric Stein-Like estimator has the advantage of allowing for a direct estimation of its shrinkage parameter. In this paper, we demonstrate how this Stein-like estimator can be implemented in the context of the estimation of natural direct and natural indirect effects of treatments in randomized controlled trials. The performance of the competing methods is studied in a simulation study, in which both the strength of hidden confounding and the strength of the instruments are independently varied. These considerations are motivated by a trial in mental health, evaluating the impact of a primary care-based intervention to reduce depression in the elderly.

scholarly journals A Two-Stage Joint Modeling Method for Causal Mediation Analysis in the Presence of Treatment Noncompliance

2020 ◽  
Vol 8 (1) ◽  
pp. 131-149
Author(s):  
Soojin Park ◽  
Esra Kürüm
Keyword(s):  
Mediation Analysis ◽  
Joint Modeling ◽  
Modeling Method ◽  
Sensitivity Analyses ◽  
Two Stage ◽  
Mediation Effects ◽  
Causal Mediation ◽  
Causal Mediation Analysis ◽  
Treatment Noncompliance ◽  
The Impact

AbstractEstimating the effect of a randomized treatment and the effect that is transmitted through a mediator is often complicated by treatment noncompliance. In literature, an instrumental variable (IV)-based method has been developed to study causal mediation effects in the presence of treatment noncompliance. Existing studies based on the IV-based method focus on identifying the mediated portion of the intention-to-treat effect, which relies on several identification assumptions. However, little attention has been given to assessing the sensitivity of the identification assumptions or mitigating the impact of violating these assumptions. This study proposes a two-stage joint modeling method for conducting causal mediation analysis in the presence of treatment noncompliance, in which modeling assumptions can be employed to decrease the sensitivity to violation of some identification assumptions. The use of a joint modeling method is also conducive to conducting sensitivity analyses to the violation of identification assumptions. We demonstrate our approach using the Jobs II data, in which the effect of job training on job seekers’ mental health is examined.

Bayesian sensitivity analysis for unmeasured confounding in causal mediation analysis

2017 ◽  
Vol 28 (2) ◽  
pp. 515-531 ◽  
Author(s):  
Lawrence C McCandless ◽  
Julian M Somers
Keyword(s):  
Sensitivity Analysis ◽  
Mediation Analysis ◽  
Indirect Effects ◽  
Direct And Indirect Effects ◽  
Unmeasured Confounding ◽  
Intermediate Variable ◽  
Causal Mediation ◽  
Important Concern ◽  
Analysis Technique ◽  
Causal Mediation Analysis

Causal mediation analysis techniques enable investigators to examine whether the effect of the exposure on an outcome is mediated by some intermediate variable. Motivated by a data example from epidemiology, we consider estimation of natural direct and indirect effects on a survival outcome. An important concern is bias from confounders that may be unmeasured. Estimating natural direct and indirect effects requires an elaborate series of assumptions in order to identify the target quantities. The analyst must carefully measure and adjust for important predictors of the exposure, mediator and outcome. Omitting important confounders may bias the results in a way that is difficult to predict. In recent years, several methods have been proposed to explore sensitivity to unmeasured confounding in mediation analysis. However, many of these methods limit complexity by relying on a handful of sensitivity parameters that are difficult to interpret, or alternatively, by assuming that specific patterns of unmeasured confounding are absent. Instead, we propose a simple Bayesian sensitivity analysis technique that is indexed by four bias parameters. Our method has the unique advantage that it is able to simultaneously assess unmeasured confounding in the mediator–outcome, exposure–outcome and exposure–mediator relationships. It is a natural Bayesian extension of the sensitivity analysis methodologies of VanderWeele, which have been widely used in the epidemiology literature. We present simulation findings, and additionally, we illustrate the method in an epidemiological study of mortality rates in criminal offenders from British Columbia.

scholarly journals Prying Open the Black Box of Causality - A Causal Mediation Analysis Test of Procedural Justice Policing

2017 ◽  
Author(s):  
Krisztián Pósch
Keyword(s):  
Procedural Justice ◽  
Treatment Effect ◽  
Mediation Analysis ◽  
Black Box ◽  
Mediation Effects ◽  
Causal Mediation ◽  
Treatment Implementation ◽  
Causal Mediation Analysis ◽  
The Impact ◽  
Sense Of Power

Objectives: Review causal mediation analysis as a method for estimating and assessing direct and indirect effects in experimental criminology. Test procedural justice theory by examining the extent to which procedural justice mediates the impact of contact with the police on various outcomes. Apply causal mediation analysis to better interpret data from a field experiment that had suffered from a particular type of implementation failure.Methods: Data from a block-randomised controlled trial of procedural justice policing (the Scottish Community Engagement Trial) were analysed. All constructs were measured using surveys distributed during roadside police checks. The treatment implementation was assessed by analysing the treatment effect consistency and heterogeneity. Causal mediation analysis and sensitivity analysis were used to assess the mediating role of procedural justice.Results: First, the treatment effect was consistent and fairly homogeneous, indicating that the systematic variation in the study is attributable to the design. Second, procedural justice acts as a mediator channelling the treatment’s effect towards normative alignment (NIE=-0.207), duty to obey (NIE=-0.153), sense of power (NIE=-0.078), and social identity (NIE=-0.052), all of which are moderately robust to unmeasured confounding. The NIEs for risk of sanction and personal morality were highly sensitive, while for coerced obligation and sense of power they were non-significant. Conclusions: Causal mediation analysis is a versatile tool that can salvage experiments with systematic yet ambiguous treatment effects by allowing researchers to “pry open” the black box of causality. Most of the theoretical propositions of procedural justice policing were supported. Future studies are needed with more discernible causal mediation effects.

Causal Mediation Analysis for Estimating Direct and Indirect Effects of an Environmental Mixture on Child Neurodevelopment

2018 ◽  
Vol 2017 (1) ◽  
pp. 433
Author(s):  
Katrina L. Devick ◽  
Jennifer F. Bobb ◽  
Maitreyi Mazumdar ◽  
Birgit Claus Henn ◽  
David C. Bellinger ◽  
...  
Keyword(s):  
Mediation Analysis ◽  
Indirect Effects ◽  
Direct And Indirect Effects ◽  
Causal Mediation ◽  
Child Neurodevelopment ◽  
Causal Mediation Analysis

THE IMPACT OF A HAUSMAN PRETEST ON THE ASYMPTOTIC SIZE OF A HYPOTHESIS TEST

2009 ◽  
Vol 26 (2) ◽  
pp. 369-382 ◽  
Author(s):  
Patrik Guggenberger
Keyword(s):  
Least Squares ◽  
Hypothesis Test ◽  
Ordinary Least Squares ◽  
Reduced Form ◽  
Structural Parameter ◽  
Ratio Test ◽  
Two Stage ◽  
Weak Instruments ◽  
Asymptotic Size ◽  
The Impact

This paper investigates the asymptotic size properties of a two-stage test in the linear instrumental variables model when in the first stage a Hausman (1978) specification test is used as a pretest of exogeneity of a regressor. In the second stage, a simple hypothesis about a component of the structural parameter vector is tested, using a t-statistic that is based on either the ordinary least squares (OLS) or the two-stage least squares estimator (2SLS), depending on the outcome of the Hausman pretest. The asymptotic size of the two-stage test is derived in a model where weak instruments are ruled out by imposing a positive lower bound on the strength of the instruments. The asymptotic size equals 1 for empirically relevant choices of the parameter space. The size distortion is caused by a discontinuity of the asymptotic distribution of the test statistic in the correlation parameter between the structural and reduced form error terms. The Hausman pretest does not have sufficient power against correlations that are local to zero while the OLS-based t-statistic takes on large values for such nonzero correlations. Instead of using the two-stage procedure, the recommendation then is to use a t-statistic based on the 2SLS estimator or, if weak instruments are a concern, the conditional likelihood ratio test by Moreira (2003).

Sign in / Sign up

Export Citation Format

Share Document