Scaled envelopes: scale-invariant and efficient estimation in multivariate linear regression
缩放包络：多元线性回归中的尺度不变与高效估计AI 解读 ↓
← 返回论文列表与筛选
阅读说明
本页整理公开书目信息，不生成学术质量评分。AI 解读根据题名及所注明的摘要或原文片段生成，未完成全文核验，可能有误；具体假设、证明与结论请核对原文。摘要保留来源文本，未经模型改写。
DOI 首次登记不等于接收或发表日期。部分更正、评论等条目没有独立摘要；缺少可靠材料时保留待补状态。
引用量采用 OpenAlex 单条关联记录，不合计多个记录；关联记录可能包含正式发表版本，不等于此版本单独引用。缺失不视为零，引用指标不代表论文质量。
关注作者按书目中的完整姓名匹配，同名和姓名变体需人工辨别。相关论文先按共享细分方向和具体方法排列，再参考通用方法与发表日期。
分类依据与纠错
AI 根据所提供的论文材料归类：摘要研究一般多元回归的包络估计，未明确涉及因子模型或其他所列专门方向。
反馈分类问题
正在核对关联版本…
代码、数据与补充材料
正在核对对应版本的资源…
关注这篇论文的作者
缩放包络：多元线性回归中的尺度不变与高效估计
- 研究问题
- 多元响应具有不同计量尺度时，如何解决原有包络方法对响应缩放不具不变性的问题，并保留提高回归系数估计效率的潜力？
- 方法与思路
- 作者提出缩放包络模型，估计适当的响应尺度，推导基于似然的估计量，并研究其在不同情形下的理论性质。
- 材料陈述的贡献
- 摘要称新模型具有尺度不变性，并能在原包络模型不带来效率收益的情形下获得显著收益；模拟和实例支持作者的理论陈述。
- 阅读关注点
- 应核对尺度参数与包络子空间的识别、维数选择、似然优化方式，以及效率比较的基准、假设和适用情形。
- 阅读边界
- 材料为版本未注明的索引摘要；未列出尺度识别条件、理论假设或效率增益数值，不能认定普遍优越或推断刊本变化。
原文摘要
Efficient estimation of the regression coefficients is a fundamental problem in multivariate linear regression. The envelope model proposed by Cook et al. (2010) was shown to have the potential to achieve substantial efficiency gains by accounting for linear combinations of the response vector that are essentially immaterial to coefficient estimation. This requires in part that the distribution of those linear combinations be invariant to changes in the nonstochastic predictor vector. However, inference based on an envelope is not invariant or equivariant under rescaling of the responses, tending to limit application to responses that are measured in the same or similar units. The efficiency gains promised by envelopes often cannot be realized when the responses are measured in different scales. To overcome this limitation and broaden the scope of envelope methods, we propose a scaled version of the envelope model, which preserves the potential of the original envelope methods to increase efficiency and is invariant to scale changes. Likelihood-based estimators are derived and theoretical properties of the estimators are studied in various circumstances. It is shown that estimating appropriate scales for the responses can produce substantial efficiency gains when the original envelope model offers none. Simulations and an example are given to support the theoretical claims.