Research map / Probabilistic and scientific modeling / Treatment effects
High Precision Causal Model Evaluation with Conditional Randomization
Chao Ma, Cheng Zhang
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 pairs estimator instead of naive IPW estimator | causal error estimation · csuite_1-3, logistic assignment | variance, bias, MSE | Improved |
| use pairs estimator instead of naive IPW under random subsampling | causal error estimation · csuite_1-3, random subsampling | variance, bias, MSE | Improved |
| increase treatment assignment imbalance (larger sigma_beta^2) for naive estimator | naive IPW estimator · csuite, logistic assignment | variance | Hurt |
| increase imbalance for pairs estimator | pairs estimator · csuite, logistic assignment | variance | No clear effect |
| pairs estimator vs RCT estimator | pairs estimator · csuite | variance, bias, MSE | No clear effect |
| use self-normalized or linearly-modified IPW instead of naive | IPW estimator · csuite | variance and bias | Mixed |
| pairs estimator on learned causal models (DML, Causal Forest, DR variants) | causal error estimation · EconML models, sigma_beta^2 = 1, 5, 10 | variance, bias, MSE | Improved |
| pairs estimator vs naive for Causal Forest | pairs estimator · Causal Forest | variance and bias | Hurt |
| pairs estimator for DML Kernel | pairs estimator · DML Kernel | variance and bias | Hurt |
| test independence of residual noise from Y^T and b (Assumption A) | pairs estimator assumptions · learned causal models | Pearson correlation and p-values | No clear effect |
Closest papers by meaning
- Efficient Randomized Experiments Using Foundation Models (NeurIPS 2025 · similarity 0.87)
- ABC3: Active Bayesian Causal Inference with Cohn Criteria in Randomized Experiments (AAAI 2025 · similarity 0.86)
- Inverse-Variance Weighting for Estimation of Heterogeneous Treatment Effects (ICML 2024 · similarity 0.86)
- Do Contemporary Causal Inference Models Capture Real-World Heterogeneity? Findings from a Large-Scale Benchmark (ICLR 2025 · similarity 0.86)
- Beyond the Average: Distributional Causal Inference under Imperfect Compliance (NeurIPS 2025 · similarity 0.85)
- Generalized Bayes for Causal Inference (NeurIPS 2026 · 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.
