Research map / Learning theory and algorithms / Sample complexity
Additive Nonlinear Quantile Regression in Ultra-high Dimension
Ben Sherwood, Adam Maidman
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 SCAD penalty instead of lasso in the quantile additive model (ablation) | QA-LASSO · Setting IA, p=300 | false positives: 14.17 -> 2.01 | Improved |
| use SCAD penalty instead of lasso in the quantile additive model (ablation) | QA-LASSO · Setting IA, p=600 | MSPE: 1.43 -> 1.07 | Improved |
| use quantile additive SCAD instead of mean additive SCAD under T3 errors | MA-SCAD · Setting IB, p=600 | false positives: 40.37 -> 3.32 | Improved |
| use quantile additive SCAD instead of mean additive SCAD under T3 errors | MA-SCAD · Setting IB, p=600 | MSPE: 4.99 -> 2.99 | Improved |
| use mean additive SCAD instead of quantile additive SCAD under homoscedastic normal errors | QA-SCAD CD · Setting IA, p=100 | MSPE: 1.05 vs 1.03 | No clear effect |
| use quantile additive SCAD instead of mean additive models for the 0.9 quantile with heteroscedastic errors | MA-SCAD · Setting IC, p=300 | true positives: 2.1 -> 3 | Improved |
| use linear programming instead of coordinate descent for QA-SCAD (ablation) | QA-SCAD CD · Setting IC, p=300 | MSPE: 2.05 -> 1.75 | Improved |
| use coordinate descent (QICD) instead of linear programming (ablation) | QA-SCAD LP · Settings IA and IB | computation time | Improved |
| use additive nonlinear quantile model instead of linear quantile model with SCAD | QL-SCAD · Setting IIC nonadditive, p=300 | MSPE: 3.54 -> 3.07 | Improved |
| use additive nonlinear quantile model instead of linear quantile model with SCAD | QL-SCAD · Setting IIA linear, p=300 | MSPE: 2.9 vs 2.91 | No clear effect |
| use additive nonlinear quantile model instead of linear quantile model with SCAD | QL-SCAD · Ground pork data | quantile bias: 0.32 -> 0.21 | Improved |
| use additive nonlinear quantile model instead of linear quantile model | QL-SCAD · TRIM32 gene expression | MAPE: 69 of 100 | Improved |
Closest papers by meaning
- Ultra-High Dimensional Single-Index Quantile Regression (JMLR 2020 · similarity 0.91)
- Non-parametric Quantile Regression via the K-NN Fused Lasso (JMLR 2021 · similarity 0.89)
- Confidence Intervals and Hypothesis Testing for High-dimensional Quantile Regression: Convolution Smoothing and Debiasing (JMLR 2023 · similarity 0.88)
- Online Quantile Regression (JMLR 2025 · similarity 0.88)
- Distributed High-Dimensional Quantile Regression: Estimation Efficiency and Support Recovery (ICML 2024 · similarity 0.86)
- Relaxed Quantile Regression: Prediction Intervals for Asymmetric Noise (ICML 2024 · similarity 0.84)
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.
