Testing Mediation Effects Using Logic of Boolean Matrices
From MaRDI portal
Abstract: Mediation analysis is becoming an increasingly important tool in scientific studies. A central question in high-dimensional mediation analysis is to infer the significance of individual mediators. The main challenge is the sheer number of possible paths that go through all combinations of mediators. Most existing mediation inference solutions either explicitly impose that the mediators are conditionally independent given the exposure, or ignore any potential directed paths among the mediators. In this article, we propose a novel hypothesis testing procedure to evaluate individual mediation effects, while taking into account potential interactions among the mediators. Our proposal thus fills a crucial gap, and greatly extends the scope of existing mediation tests. Our key idea is to construct the test statistic using the logic of Boolean matrices, which enables us to establish the proper limiting distribution under the null hypothesis. We further employ screening, data splitting, and decorrelated estimation to reduce the bias and increase the power of the test. We show our test can control both the size and false discovery rate asymptotically, and the power of the test approaches one, meanwhile allowing the number of mediators to diverge to infinity with the sample size. We demonstrate the efficacy of our method through both simulations and a neuroimaging study of Alzheimer's disease.
Recommendations
- Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators
- Estimation and inference for the indirect effect in high-dimensional linear mediation models
- A Multiple-Testing Procedure for High-Dimensional Mediation Hypotheses
- Variance component tests of multivariate mediation effects under composite null hypothesis
- High dimensional mediation analysis with latent variables
Cites work
- 10.1162/153244303321897717
- A general theory of hypothesis tests and confidence regions for sparse high dimensional models
- A knockoff filter for high-dimensional selective inference
- Confidence intervals for low dimensional parameters in high dimensional linear models
- Constrained likelihood for reconstructing a directed acyclic Gaussian graph
- Double/debiased machine learning for treatment and structural parameters
- Estimating the effect of joint interventions from observational data in sparse high-dimensional settings
- Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
- Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators
- Identifiability of Gaussian structural equation models with equal error variances
- Joint significance tests for mediation effects of socioeconomic adversity on adiposity via epigenetics
- Linear hypothesis testing for high dimensional generalized linear models
- Nearly unbiased variable selection under minimax concave penalty
- On High-Dimensional Constrained Maximum Likelihood Inference
- Regression Shrinkage and Selection via The Lasso: A Retrospective
- Simultaneous analysis of Lasso and Dantzig selector
- Sure independence screening for ultrahigh dimensional feature space. With discussion and authors' reply
- The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
- The control of the false discovery rate in multiple testing under dependency.
- Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
- \(\ell_{0}\)-penalized maximum likelihood for sparse directed acyclic graphs
- \(p\)-values for high-dimensional regression
Cited in
(15)- Self-normalized Cramér type moderate deviation theorem for Gaussian approximation
- Mediation analysis for high-dimensional mediators and outcomes with an application to multimodal imaging data
- Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators
- Multivariate dynamic mediation analysis under a reinforcement learning framework
- De-confounding Causal Inference Using Latent Multiple-Mediator Pathways
- Genome-wide analyses of sparse mediation effects under composite null hypotheses
- Estimation and inference for the indirect effect in high-dimensional linear mediation models
- Variance component tests of multivariate mediation effects under composite null hypothesis
- Time-varying mediation analysis for incomplete data with application to DNA methylation study for PTSD
- Discovery and Inference of a Causal Network with Hidden Confounding
- On optimal two‐stage testing of multiple mediators
- Causal mediation analysis: selection with asymptotically valid inference
- Adaptive-to-sub-null testing for mediation effects in structural equation models
- A Multiple-Testing Procedure for High-Dimensional Mediation Hypotheses
- Sequential pathway inference for multimodal neuroimaging analysis
This page was built for publication: Testing Mediation Effects Using Logic of Boolean Matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6110717)
