Maximum-Variance-Reduction Stratification for Improved Subsampling
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915981737066496 |
|---|---|
| author | Wang, Dingyi Wang, Haiying Hu, Qingpei |
| author_facet | Wang, Dingyi Wang, Haiying Hu, Qingpei |
| contents | Subsampling is a widely used and effective approach for addressing the computational challenges posed by massive datasets. Substantial progress has been made in developing non-uniform, probability-based subsampling schemes that prioritize more informative observations. We propose a novel stratification mechanism that can be combined with existing subsampling designs to further improve estimation efficiency. We establish the estimator's asymptotic normality and quantify the resulting efficiency gains, which enables a principled procedure for selecting stratification variables and interval boundaries that target reductions in asymptotic variance. The resulting algorithm, Maximum-Variance-Reduction Stratification (MVRS), achieves significant improvements in estimation efficiency while incurring only linear additional computational cost. MVRS is applicable to both non-uniform and uniform subsampling methods. Experiments on simulated and real datasets confirm that MVRS markedly reduces estimator variance and improves accuracy compared with existing subsampling methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_18075 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Maximum-Variance-Reduction Stratification for Improved Subsampling Wang, Dingyi Wang, Haiying Hu, Qingpei Methodology Subsampling is a widely used and effective approach for addressing the computational challenges posed by massive datasets. Substantial progress has been made in developing non-uniform, probability-based subsampling schemes that prioritize more informative observations. We propose a novel stratification mechanism that can be combined with existing subsampling designs to further improve estimation efficiency. We establish the estimator's asymptotic normality and quantify the resulting efficiency gains, which enables a principled procedure for selecting stratification variables and interval boundaries that target reductions in asymptotic variance. The resulting algorithm, Maximum-Variance-Reduction Stratification (MVRS), achieves significant improvements in estimation efficiency while incurring only linear additional computational cost. MVRS is applicable to both non-uniform and uniform subsampling methods. Experiments on simulated and real datasets confirm that MVRS markedly reduces estimator variance and improves accuracy compared with existing subsampling methods. |
| title | Maximum-Variance-Reduction Stratification for Improved Subsampling |
| topic | Methodology |
| url | https://arxiv.org/abs/2601.18075 |