Regression with stagewise minimization on risk function
Regression with stagewise minimization on risk function
复制标题
风险函数阶段最小化回归
DOI：
10.1016/j.csda.2018.12.011
复制
发表时间：
2019
影响因子：
1.8
通讯作者：
Takuma Yoshida and Kanta Naito
中科院分区：
文献类型：
--
作者：
Nazarov Anatoly;Phung-Duc Tuan;Paul Svetlana;Lizyura Olga;Shulgina Kseniya;Takuma Yoshida and Kanta Naito
本文研究了一种基于经验风险最小化的曲线估计。估计器由字典中单词（学习器）的凸组合组成。在所提出的阶段式算法的每一步中都会选择一个单词，从而最小化一定的散度度量。开发了估计器的非渐近误差界，并且表明随着算法迭代次数的增加，误差界变得尖锐。模拟研究和真实数据示例证实了估计器的性能。
This paper studies a curve estimation based on empirical risk minimization. The estimator is composed as a convex combination of words (learners) in a dictionary. A word is selected in each step of the proposed stagewise algorithm, which minimizes a certain divergence measure. A non-asymptotic error bound of the estimator is developed, and it is shown that the error bound becomes sharp as the number of iterations of the algorithm increases. A simulation study and real data example confirm the performance of the estimator.