Research map / Probabilistic and scientific modeling / Treatment effects
Learning Treatment Effects in Panels with General Intervention Patterns
Vivek Farias, Andrew A Li, Tianyi Peng
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 |
|---|---|---|---|
| use De-biased Convex instead of SDID | SDID · Tobacco block pattern | normalized error: 0.23 -> 0.15 | Improved |
| use De-biased Convex instead of SDID | SDID · Tobacco stagger pattern | normalized error: 0.16 -> 0.10 | Improved |
| use De-biased Convex instead of MC-NNM | MC-NNM · Tobacco block pattern | normalized error: 0.27 -> 0.15 | Improved |
| use De-biased Convex instead of RSC | RSC · Tobacco stagger pattern | normalized error: 0.20 -> 0.10 | Improved |
| use De-biased Convex instead of OLS | OLS · Tobacco block pattern | normalized error: 0.38 -> 0.15 | Improved |
| use De-biased Convex instead of MC-NNM on adaptive treatment pattern | MC-NNM · Sales adaptive pattern | normalized error: 0.13 -> 0.02 | Improved |
| use De-biased Convex instead of OLS on adaptive treatment pattern | OLS · Sales adaptive pattern | normalized error: 0.20 -> 0.02 | Improved |
| use treated observations in estimation (De-biased Convex vs MC-NNM which ignores them) (ablation) | MC-NNM · Sales adaptive pattern | normalized error: 27.6% -> 5.3% | Improved |
| use De-biased Convex instead of MC-NNM on real data | MC-NNM · Rossmann real data | test error: 0.07 -> 0.04 | Improved |
| use De-biased Convex instead of OLS on real data | OLS · Rossmann real data | test error: 0.18 -> 0.04 | Improved |
Closest papers by meaning
- It’s Hard to Be Normal: The Impact of Noise on Structure-agnostic Estimation (NeurIPS 2025 · similarity 0.86)
- A Meta-learner for Heterogeneous Effects in Difference-in-Differences (ICML 2025 · similarity 0.85)
- Beyond the Average: Distributional Causal Inference under Imperfect Compliance (NeurIPS 2025 · similarity 0.85)
- Targeted Synthetic Control Method (NeurIPS 2026 · similarity 0.85)
- Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models (NeurIPS 2026 · similarity 0.85)
- Active Treatment Effect Estimation via Limited Samples (ICML 2025 · similarity 0.85)
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.
