Research map / Probabilistic and scientific modeling / Treatment effects
Mutual Information Based Matching for Causal Inference with Observational Data
Lei Sun, Alexander G. Nikolaev
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 MIM-Marginal MIP instead of propensity score matching | matching · Synthetic 25c10k, 25 covariates | ATT estimate (true 0): -14.753 -> 12.170 | Improved |
| use marginal-bin MIM instead of joint-bin MIM | MIM · Synthetic 25c10k, 10 covariates | ATT estimate (true 0): 16.646 -> -1.123 | Improved |
| use MIM-Joint with a single covariate | MIM-Joint · Synthetic 25c10k, 1 covariate | ATT estimate (true 0): 0.006 -> 0.006 | No clear effect |
| use MIM-Marginal (MIP) instead of BOSS | matching · Synthetic 25c10k, 25 covariates | ATT estimate (true 0): 6.618 -> 12.170 | Hurt |
| use MIM-Marginal (Sequential) instead of BOSS | matching · Synthetic 25c10k, 25 covariates | ATT estimate (true 0): 6.618 -> 8.448 | Mixed |
| use MIM-Marginal (MIP) instead of Mahalanobis matching | matching · Synthetic 25c10k, 10 covariates | KS average: 0.112 -> 0.070 | Improved |
| use MIM-Marginal (MIP) instead of propensity score matching | matching · Synthetic 25c10k, 10 covariates | KS average: 0.154 -> 0.034 | Improved |
| increase the number of covariates from 10 to 25 for MIM-Marginal (Sequential) | MIM-Marginal Sequential · Synthetic 25c10k | ATT estimate (true 0): 0.517 -> 8.448 | Hurt |
| lower mutual information of the control group | MIM-selected control group · Synthetic 25c10k, 10 covariates | ATT estimate: MI <0.002 to [0.07,0.08) | Improved |
| remove the fixed control group size restriction | MIM-Marginal · NSW LaLonde | MI: 0.002383 -> 0.001824 | Improved |
Closest papers by meaning
- Matching without Group Barrier for Heterogeneous Treatment Effect Estimation (ICLR 2026 · similarity 0.87)
- Causal Aggregation: Estimation and Inference of Causal Effects by Constraint-Based Data Fusion (JMLR 2022 · similarity 0.87)
- Hierarchical and Density-based Causal Clustering (NeurIPS 2024 · similarity 0.86)
- Automating the Selection of Proxy Variables of Unmeasured Confounders (ICML 2024 · similarity 0.86)
- Doubly robust identification of treatment effects from multiple environments (ICLR 2025 · similarity 0.85)
- Data Fusion for Partial Identification of Causal Effects (NeurIPS 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.
