Research map / Probabilistic and scientific modeling / Treatment effects
Staggered Rollout Designs Enable Causal Inference Under Interference Without Network Knowledge
Mayleen Cortez, Matthew Eichhorn, Christina 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 |
|---|---|---|---|
| increase population size n | Staggered rollout PI estimator (CRD) · Simulated networks, linear and polynomial potential outcomes | estimator variance | Improved |
| increase maximum treated proportion k/n | Staggered rollout PI estimator (CRD) · Simulated networks | estimator variance | Improved |
| use staggered rollout PI estimator instead of baseline TTE estimators | TTE estimation under network interference · Simulated networks | relative bias | Improved |
| increase ratio r of network to direct effects | Baseline TTE estimators · Simulated networks | relative bias | Hurt |
| set ratio r close to 0 (no network effects) | All TTE estimators · Simulated networks | relative bias | No clear effect |
| use realized treatment counts k-hat/n instead of expected fraction k/n for interpolation (ablation) | Staggered rollout PI estimator · Simulated networks | relative bias and variance | No clear effect |
| interpolate on realized treatment fraction k-hat/n instead of expected fraction p under Bernoulli design (ablation) | Staggered rollout PI estimator · Simulated networks, Bernoulli design | estimator variance | Improved |
| use completely randomized design instead of Bernoulli design for baselines | Benchmark estimators · Simulated networks | balance of design | Improved |
| naive regression with homogeneous effects (LS) instead of the PI estimator | Least squares estimators · Simulated networks with heterogeneity | relative bias | Hurt |
Closest papers by meaning
- Independent-Set Design of Experiments for Estimating Treatment and Spillover Effects under Network Interference (ICLR 2024 · similarity 0.91)
- Higher-Order Causal Message Passing for Experimentation with Complex Interference (NeurIPS 2024 · similarity 0.90)
- Online Experimental Design With Estimation-Regret Trade-off Under Network Interference (NeurIPS 2025 · similarity 0.88)
- Telling Peer Direct Effects from Indirect Effects in Observational Network Data (ICML 2025 · similarity 0.86)
- Active Treatment Effect Estimation via Limited Samples (ICML 2025 · similarity 0.86)
- Just Ramp-Up: Debiasing Regression-based Estimator for A/B Tests under Network Interference (NeurIPS 2026 · similarity 0.86)
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.
