University of Bristol(布里斯托尔大学);The University of Edinburgh(爱丁堡大学);Maxwell Institute for Mathematical Sciences(麦克斯韦数学科学研究所);University of Cambridge(剑桥大学)
Henry W. J. Reeve, Timothy I. Cannings, Richard J. Samworth
备注:
65 pages, 2 figures, to appear in the Annals of Statistics
本文研究回归设定下识别回归函数超过阈值的特征空间区域的最优子群选择问题,将其形式化为约束优化,并确定了遗憾在样本量和第 I 类错误概率方面的极小极大最优速率,同时推广到异质性处理效应设定。
摘要:
在临床试验和其他应用中,我们经常看到特征空间中似乎表现出有趣行为的区域,但不清楚这些观察到的现象是否在总体层面上得到反映。聚焦于回归设定,我们考虑子群选择挑战,即识别特征空间中回归函数超过预定阈值的一个区域。我们将该问题表述为约束优化问题,其中我们寻求一个低复杂度、数据依赖的选择集,在该集合上,以有保证的概率,回归函数一致地至少与阈值一样大;在此约束下,我们希望该区域在边际特征分布下包含尽可能多的质量。这引出了一个自然的遗憾概念,我们的主要贡献是确定该遗憾在样本量和第 I 类错误概率方面的极小极大最优速率。该速率涉及控制回归函数平滑性的参数与量化总体层面最优选择集可被良态子集族逼近程度的指数之间的精细相互作用。最后,我们通过说明这些结果如何推广到处理和对照设定(其中兴趣在于异质性处理效应)来扩展我们先前结果的范围。
英文摘要:
In clinical trials and other applications, we often see regions of the feature space that appear to exhibit interesting behaviour, but it is unclear whether these observed phenomena are reflected at the population level. Focusing on a regression setting, we consider the subgroup selection challenge of identifying a region of the feature space on which the regression function exceeds a pre-determined threshold. We formulate the problem as one of constrained optimisation, where we seek a low-complexity, data-dependent selection set on which, with a guaranteed probability, the regression function is uniformly at least as large as the threshold; subject to this constraint, we would like the region to contain as much mass under the marginal feature distribution as possible. This leads to a natural notion of regret, and our main contribution is to determine the minimax optimal rate for this regret in both the sample size and the Type I error probability. The rate involves a delicate interplay between parameters that control the smoothness of the regression function, as well as exponents that quantify the extent to which the optimal selection set at the population level can be approximated by families of well-behaved subsets. Finally, we expand the scope of our previous results by illustrating how they may be generalised to a treatment and control setting, where interest lies in the heterogeneous treatment effect.
Efficient Subgroup Analysis via Optimal Trees with Global Parameter Fusion
Division of Biostatistics, University of California, Berkeley(加州大学伯克利分校生物统计学系);Helen Wills Neuroscience Institute, University of California, Berkeley(加州大学伯克利分校海伦·威尔斯神经科学研究所)
Data driven extreme value distribution estimation: Derivation of the Mean Integrated Squared Error, optimal bandwidth selection and stability conditions
Optimal Feature Selection in High-Dimensional Discriminant Analysis
我们考虑高维判别分析问题。对于这个问题,已经提出了不同的方法,并通过建立分类风险的精确收敛速度以及判别规则的l2收敛结果来证明。然而,尖锐的理论分析,这些程序的变量选择性能尚未建立,即使模型解释是科学数据分析的根本重要性。本文通过使用稀疏判别分析(Mai et al.,2012年)。通过仔细的分析,我们建立的收敛速度显着快于最知名的结果,并承认一个最佳的缩放的样本量n,维数p,和稀疏水平s的高维设置。在高维判别分析的背景下,变量选择问题的充分条件的补充必要的信息理论的限制。利用数值等价结果,我们的方法还建立了ROAD估计量的最佳结果(Fan等人,2012)和稀疏最优缩放估计器(Clemmensen等人,2011年)。此外,我们分析了一个穷举搜索过程,其性能作为一个基准,并表明它是变量选择一致的条件下较弱。还提供了大量的模拟,证明边界的清晰度。