Research map / Probabilistic and scientific modeling / Causal discovery
Conditional Independence Testing with Heteroskedastic Data and Applications to Causal Discovery
Wiebke Günther, Urmi Ninad, Jonas Wahl, Jakob Runge
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Replace OLS residual regression with WLS using known or estimated variance weights | ParCorr partial correlation CI test · Simulated heteroskedastic CI testing | p-value calibration (KS statistic) | Improved |
| Replace OLS residual regression with WLS using known or estimated variance weights | ParCorr partial correlation CI test · Simulated heteroskedastic CI testing | AUPC | Improved |
| Use ParCorr-WLS instead of ParCorr-OLS | PC algorithm · Simulated causal discovery with increasing heteroskedasticity | false positive rate | Improved |
| Use ParCorr-WLS instead of ParCorr-OLS | PC algorithm · Simulated causal discovery with increasing heteroskedasticity | true positive rate | Improved |
| Use ParCorr-WLS instead of ParCorr-OLS | ParCorr partial correlation CI test · Homoskedastic CI testing | AUPC | No clear effect |
| Estimate weights with the window approach instead of using ground-truth weights (ablation) | ParCorr-WLS · Simulated heteroskedastic CI testing | overall performance | No clear effect |
| Use ParCorr-OLS instead of ParCorr-WLS | ParCorr partial correlation CI test · Simulated heteroskedastic CI testing | AUPC | Hurt |
| Use ParCorr-OLS instead of ParCorr-WLS when both tested variables have heteroskedasticity | ParCorr partial correlation CI test · Simulated heteroskedastic CI testing | p-value calibration (KS statistic) | Hurt |
| Use ParCorr-OLS instead of ParCorr-WLS | PC algorithm · Simulated causal discovery with increasing heteroskedasticity | false positive rate | Hurt |
| Use CMIknn instead of ParCorr-WLS | Conditional independence testing · Simulated heteroskedastic CI testing | detection power | Hurt |
Closest papers by meaning
- Causal Discovery via Conditional Independence Testing with Proxy Variables (ICML 2024 · similarity 0.91)
- PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data (NeurIPS 2026 · similarity 0.89)
- On the sample complexity of conditional independence testing with Von Mises estimator with application to causal discovery (ICML 2024 · similarity 0.89)
- A Conditional Independence Test in the Presence of Discretization (ICLR 2025 · similarity 0.87)
- A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery (ICLR 2025 · similarity 0.87)
- Signature Kernel Conditional Independence Tests in Causal Discovery for Stochastic Processes (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
