Research map / Probabilistic and scientific modeling / Causal discovery
Power-Boosted Granger-Causal Discovery for Large Heterogeneous Panel Data
Yiheng Gu, Xiufan Yu
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 |
|---|---|---|---|
| add the power enhancement component J_PE to the pivotal test | DH/HPJ pivotal test · Type I error, all (T,N) settings | empirical Type I error | No clear effect |
| add the power enhancement component J_PE to the pivotal test | DH, T=50 N=500 · Type I error | empirical Type I error: 0.187 -> 0.052 | Improved |
| use HPJ instead of DH when N >= T-K (data-adaptive pivotal choice) (ablation) | PE-PGCT · Type I error, T=100 N=5 vs HPJ | empirical Type I error: 0.189 -> 0.044 | Improved |
| add J_PE to the pivotal test | HPJ under sparse alternatives (N0=1) · power vs signal strength | power | Improved |
| add J_PE to the pivotal test | DH under sparse alternatives (N0=1) · power vs signal strength | power | Improved |
| add J_PE to the pivotal test when the alternative becomes dense | DH, increasing number of signal units N0 · power vs signal sparsity | power | Mixed |
| apply PE-PGCT instead of the pivotal test to bank data | Large banks, full period · p-value Granger causality, FDIC panel | p-value: 9.3 × 10−10 -> < 10−16 | Improved |
| apply PE-PGCT instead of the pivotal test to bank data | Small banks, DFA · p-value Granger causality, FDIC panel | p-value: 4.5 × 10−12 -> < 10−16 | Improved |
| apply PE-PGCT instead of the pivotal test to bank data | Large banks, Basel II and DFA · p-value Granger causality, FDIC panel | p-value: 0.011 -> 0.011 | No clear effect |
Closest papers by meaning
- Exploring Neural Granger Causality with xLSTMs: Unveiling Temporal Dependencies in Complex Data (NeurIPS 2025 · similarity 0.84)
- Jacobian Regularizer-based Neural Granger Causality (ICML 2024 · similarity 0.83)
- CUTS+: High-Dimensional Causal Discovery from Irregular Time-Series (AAAI 2024 · similarity 0.83)
- Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models (NeurIPS 2026 · similarity 0.82)
- Identification of Causal Structure with Latent Variables Based on Higher Order Cumulants (AAAI 2024 · similarity 0.82)
- A Practical Approach to Causal Inference over Time (AAAI 2025 · similarity 0.82)
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.
