Nonparametric conditional local independence testing
From MaRDI portal
Abstract: Conditional local independence is an asymmetric independence relation among continuous time stochastic processes. It describes whether the evolution of one process is directly influenced by another process given the histories of additional processes, and it is important for the description and learning of causal relations among processes. We develop a model-free framework for testing the hypothesis that a counting process is conditionally locally independent of another process. To this end, we introduce a new functional parameter called the Local Covariance Measure (LCM), which quantifies deviations from the hypothesis. Following the principles of double machine learning, we propose an estimator of the LCM and a test of the hypothesis using nonparametric estimators and sample splitting or cross-fitting. We call this test the (cross-fitted) Local Covariance Test ((X)-LCT), and we show that its level and power can be controlled uniformly, provided that the nonparametric estimators are consistent with modest rates. We illustrate the theory by an example based on a marginalized Cox model with time-dependent covariates, and we show in simulations that when double machine learning is used in combination with cross-fitting, then the test works well without restrictive parametric assumptions.
Recommendations
- Testing Conditional Independence Restrictions
- Conditional Independence Specification Testing for Dependent Processes with Local Polynomial Quantile Regression
- Testing conditional independence via empirical likelihood
- Strongly consistent nonparametric tests of conditional independence
- The Hardness of Conditional Independence Testing and the Generalised Covariance Measure
Cites work
- scientific article; zbMATH DE number 1515832 (Why is no real title available?)
- scientific article; zbMATH DE number 6982970 (Why is no real title available?)
- scientific article; zbMATH DE number 7370587 (Why is no real title available?)
- A general dynamical statistical model with causal interpretation
- Brownian motion. An introduction to stochastic processes. With contributions by Björn Böttcher
- Composable Markov processes
- Conditional Independence Testing in Hilbert Spaces with Applications to Functional Data Analysis
- Double/debiased machine learning for treatment and structural parameters
- Dynamic modelling and causality
- Graphical Models for Composable Finite Markov Processes
- Graphical Models for Marked Point Processes Based on Local Independence
- Graphical modeling of stochastic processes driven by correlated noise
- Investigating Causal Relations by Econometric Models and Cross-spectral Methods
- Markov equivalence of marginalized local independence graphs
- Minimax optimal conditional independence testing
- Penalized solutions to functional regression problems
- Point processes and queues. Martingale dynamics
- Statistical modeling of causal effects in continuous time
- Statistical models based on counting processes
- The Hardness of Conditional Independence Testing and the Generalised Covariance Measure
- Uncovering causality from multivariate Hawkes integrated cumulants
Cited in
(7)- Conditional Independence Specification Testing for Dependent Processes with Local Polynomial Quantile Regression
- scientific article; zbMATH DE number 4182624 (Why is no real title available?)
- Weak equivalence of local independence graphs
- A NONPARAMETRIC HELLINGER METRIC TEST FOR CONDITIONAL INDEPENDENCE
- Double robust conditional independence test for novel biomarkers given established risk factors with survival data
- Semi-supervised distribution learning
- Strongly consistent nonparametric tests of conditional independence
This page was built for publication: Nonparametric conditional local independence testing
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6183775)
