Research map / Probabilistic and scientific modeling / Treatment effects
A Semi-parametric Estimation of Personalized Dose-response Function Using Instrumental Variables
Wei Luo, Yeying Zhu, Xuekui Zhang, Lin Lin
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 SPIVE instead of linear sisVIVE | personalized dose-response estimation · Simulation Model 3 (p=10) | D (lower is better): .544 -> .145 | Improved |
| use SPIVE instead of linear sisVIVE | personalized dose-response estimation · Simulation Model 6 (p=10) | D (lower is better): .839 -> .157 | Improved |
| use SPIVE instead of linear sisVIVE | personalized dose-response estimation · Simulation Model 1 (p=10, linear truth) | D (lower is better): .051 -> .097 | Hurt |
| raise dimension from p=10 to large p (ablation) | SPIVE · Simulation Model 1 | D (lower is better): .097 -> .104 | No clear effect |
| raise dimension from p=10 to large p (ablation) | sisVIVE · Simulation Model 1 | D (lower is better): .051 -> .198 | Hurt |
| use SMIVE instead of sisVIVE for marginal dose-response | marginal dose-response estimation · Simulation Model 4 (p=10) | D (lower is better): .660 -> .239 | Improved |
| use SMIVE instead of sisVIVE for marginal dose-response | marginal dose-response estimation · Simulation Model 1 (p=10) | D (lower is better): .048 -> .079 | Hurt |
| use SMIVE instead of sisVIVE at large p | marginal dose-response estimation · Simulation Model 1 (large p) | D (lower is better): .930 -> .084 | Improved |
| use SCAD-penalized selection instead of lasso (sisVIVE) | invalid instrument selection · Simulation Model 1 (p=10) | avg misspecified invalid instruments: .119 -> .012 | Improved |
| use SCAD-penalized selection instead of lasso (sisVIVE) | invalid instrument selection · Simulation Model 1 (large p) | avg misspecified invalid instruments: 23.3 -> 0 | Improved |
| use SCAD-penalized selection instead of lasso (sisVIVE) | invalid instrument selection · Simulation Model 6 (p=10) | avg misspecified invalid instruments: .024 -> .107 | Hurt |
| change X distribution from Gaussian (Model 4) to uniform (Model 5) (ablation) | SPIVE · Simulation (p=10) | D (lower is better): .118 -> .125 | No clear effect |
Closest papers by meaning
- Control Function Instrumental Variable Estimation of Nonlinear Causal Effect Models (JMLR 2016 · similarity 0.87)
- A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding (NeurIPS 2024 · similarity 0.84)
- Data-Driven Selection of Instrumental Variables for Additive Nonlinear, Constant Effects Models (ICML 2025 · similarity 0.84)
- Conditional Instrumental Variable Regression with Representation Learning for Causal Inference (ICLR 2024 · similarity 0.84)
- Estimating Heterogeneous Treatment Effects by Combining Weak Instruments and Observational Data (NeurIPS 2024 · similarity 0.83)
- Learning Representations of Instruments for Partial Identification of Treatment Effects (ICML 2025 · similarity 0.83)
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.
