线性模型中用于定向错误发现率控制的响应引导 knockoffs
Response-guided knockoffs for directional FDR control in linear models
- School of Mathematics and Statistics, University of Sydney（悉尼大学数学与统计学院）
- School of Mathematical and Physical Sciences, Macquarie University（麦考瑞大学数理科学学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究针对线性模型特征选择的定向 FDR 控制问题，提出响应引导 knockoff 过滤器，在更弱样本量要求下实现定向 FDR 控制，经模拟与 HIV 耐药实验验证功效优于现有方法。
AI中文摘要：
我们考虑有限样本下线性模型中的特征选择问题，需控制错误发现率（FDR）。现有基于 knockoff 的方法虽能控制定向 FDR（惩罚符号估计错误），但不针对预先指定方向的发现，且其 knockoff 构造完全与响应无关。我们提出响应引导 knockoff 过滤器，利用带噪声扰动的响应版本引导 knockoff 构造朝向可能具有目标符号的特征，同时可证明能控制定向 FDR。该方法在更弱的样本量要求 $n > p + 2$ 下运行，而现有固定-X 生成器要求 $n \geq 2p$。模拟实验与 HIV 耐药性实验表明，该方法比现有方法具有更高的功效。
英文摘要：
We consider the problem of feature selection in linear models with finite-sample control of the false discovery rate (FDR). While existing knockoff-based methods control the directional FDR, which penalises incorrect sign estimates, they do not target discoveries in a pre-specified direction, and their knockoff constructions are entirely response-agnostic. We introduce the response-guided knockoff filter, which leverages a noise-perturbed version of the response to guide knockoff construction toward features likely to have the target sign, while provably controlling the directional FDR. The method operates under a weaker sample-size requirement $n > p + 2$, compared to $n \geq 2p$ required by existing fixed-X generators. Simulations and HIV drug resistance experiments demonstrate power gains over existing methods.
