{
  "abstract": "Online controlled experiments, or A/B tests, are widely used to estimate causal effects on digital platforms. A central challenge is to improve experimental sensitivity, or statistical power, without increasing the experimental sample size. Stratified sampling is a classical variance reduction technique; however, its effectiveness depends critically on how the strata are constructed. We thus propose an optimization-based stratification framework for stratified sampling using optimal multi-way decision trees. Our method, called Optimal Multi-way Stratification Trees (OMST), formulates stratification as a path-selection problem over a feature graph. The selected paths define interpretable stratification rules and are optimized using an exact variance-minimizing binary optimization formulation under continuous proportional allocation and a Neyman-type optimal allocation. We incorporate supervised optimal binning to generate outcome-relevant candidate splits for numerical features. Furthermore, we introduce reduction procedures for redundant candidate paths and assignment constraints, substantially reducing the optimization problem size. Experiments on both a real-world and a simulated dataset demonstrate that OMST achieves comparable or superior variance reduction to existing methods while maintaining shallow and interpretable stratification trees.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2609.23308",
  "author_total": 3,
  "authors": [
    {
      "name": "Tomoka Takei"
    },
    {
      "name": "Shunnosuke Ikeda"
    },
    {
      "name": "Yuichi Takano"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/6s3g7s0jzb4hg.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2609.23308",
  "fulltext_url": null,
  "id": "6s3g7s0jzb4hg",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2609.23308",
  "language": null,
  "licence": "LicenseRef-arXiv-nonexclusive-1.0",
  "licence_label": "arXiv licence",
  "licence_source": "article",
  "licence_url": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.LG"
    ],
    [
      "arxiv",
      "stat.CO"
    ],
    [
      "openalex",
      "1803|Optimal Experimental Design Methods"
    ]
  ],
  "openalex": "W7214027253",
  "pdf_url": "https://arxiv.org/pdf/2609.23308",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/6s3g7s0jzb4hg.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:23:33Z",
      "native_id": "2609.23308",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2609.23308"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.CO",
      "subfield": "statistics"
    }
  ],
  "title": "Optimal Multi-way Decision Trees for Stratified Sampling in Online Controlled Experiments",
  "title_text": "Optimal Multi-way Decision Trees for Stratified Sampling in Online Controlled Experiments",
  "type": "preprint",
  "updated_at": "2026-10-30T12:23:33Z",
  "url": "https://research.draconitic.com/paper/6s3g7s0jzb4hg",
  "venue": "arXiv",
  "versions": []
}