Research map / Probabilistic and scientific modeling / Molecules and proteins
Regularized Estimation and Testing for High-Dimensional Multi-Block Vector-Autoregressive Models
Jiahe Lin, George Michailidis
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 |
|---|---|---|---|
| raise sample size from T=200 to T=500 when rho(A)=rho(C)=0.8 | Multi-block VAR MLE · Simulation C.3 vs C.3' | relative error of A-hat: 0.18 -> 0.13 | Improved |
| raise sample size from T=200 to T=500 when rho(A)=rho(C)=0.8 | Multi-block VAR MLE · Simulation C.3 vs C.3' | relative error of C-hat: 0.14 -> 0.06 | Improved |
| increase p1 from 100 to 200 (p2=50) | Multi-block VAR MLE · Simulation A.2 vs A.3 | relative error of A-hat: 0.38 -> 0.87 | Hurt |
| increase p1 from 100 to 200 (p2=50) | Multi-block VAR MLE · Simulation A.2 vs A.3 | relative error of B-hat: 0.31 -> 0.54 | Hurt |
| increase p1 from 50 to 100 and p2 from 20 to 50 | Multi-block VAR MLE · Simulation A.1 vs A.2 | relative error of B-hat: 0.11 -> 0.31 | Hurt |
| increase rank of B from 5 to 20 (p1=100,p2=50) | Multi-block VAR MLE · Simulation A.2 vs B.2 | relative error of B-hat: 0.31 -> 0.12 | Improved |
| increase rho(A) from 0.5 to 0.8 (p1=50,p2=20) | Multi-block VAR MLE · Simulation A.1 vs C.1 | relative error of A-hat: 0.34 -> 0.25 | Improved |
| increase rho(C) from 0.5 to 0.8 (p1=50,p2=20) | Multi-block VAR MLE · Simulation A.1 vs C.2 | relative error of C-hat: 0.15 -> 0.04 | Improved |
| replace Gaussian noise with t(df=3) noise | Multi-block VAR MLE · Simulation A.2 | relative error of A-hat: 0.38 -> 0.60 | Hurt |
| replace Gaussian noise with t(df=3) noise | Multi-block VAR MLE · Simulation B.1 | relative error of A-hat: 0.37 -> 0.61 | Hurt |
| replace Gaussian noise with t(df=3) noise | Multi-block VAR MLE · Simulation A.2 | sensitivity of A-hat: 0.97 -> 0.99 | No clear effect |
| replace Gaussian noise with elliptical noise | Multi-block VAR MLE · Simulation A.2 | relative error of B-hat: 0.31 -> 0.34 | No clear effect |
Closest papers by meaning
- High Dimensional Forecasting via Interpretable Vector Autoregression (JMLR 2020 · similarity 0.87)
- A Practical Approach to Causal Inference over Time (AAAI 2025 · similarity 0.85)
- A Set-Sequence Model for Time Series (NeurIPS 2026 · similarity 0.84)
- Jacobian Regularizer-based Neural Granger Causality (ICML 2024 · similarity 0.83)
- Discovering Mixtures of Structural Causal Models from Time Series Data (ICML 2024 · similarity 0.82)
- Flow based approach for Dynamic Temporal Causal models with non-Gaussian or Heteroscedastic Noises (NeurIPS 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.
