arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2403.17481stat.ME
收藏

一类非线性Fréchet回归

A Type of Nonlinear Fréchet Regressions

Lu Lin, Ze Chen

AI总结:

本文提出一类非线性Fréchet回归,用于在一般度量空间中拟合本质非线性模型并在Hilbert空间中唯一识别非线性结构,具有估计一致性等理论性质且表现显著优于竞争方法。

摘要:

现有的Fréchet回归实际上是在线性框架内定义的,因为Fréchet目标函数中的权重函数是线性定义的,当随机对象属于Hilbert空间时,所得的Fréchet回归函数被识别为线性模型。即使对于通常是非线性的非参数和半参数Fréchet回归,现有方法也通过局部线性(或局部多项式)技术进行处理,所得的Fréchet回归同样是(局部)线性的。我们在本文中引入了一类非线性Fréchet回归。这样的框架可用于拟合一般度量空间中本质非线性的模型,并在Hilbert空间中唯一识别非线性结构。特别地,其广义线性形式可以通过权重函数的特殊选择回到标准线性Fréchet回归。此外,由于欧几里得变量和度量空间元素是完全可分的,广义线性形式具有方法学和计算上的简便性。系统地建立了非线性Fréchet回归的有利理论性质(例如估计一致性和表示定理)。综合模拟研究和人类死亡率数据分析表明,新策略显著优于竞争方法。

英文摘要:

The existing Fréchet regression is actually defined within a linear framework, since the weight function in the Fréchet objective function is linearly defined, and the resulting Fréchet regression function is identified to be a linear model when the random object belongs to a Hilbert space. Even for nonparametric and semiparametric Fréchet regressions, which are usually nonlinear, the existing methods handle them by local linear (or local polynomial) technique, and the resulting Fréchet regressions are (locally) linear as well. We in this paper introduce a type of nonlinear Fréchet regressions. Such a framework can be utilized to fit the essentially nonlinear models in a general metric space and uniquely identify the nonlinear structure in a Hilbert space. Particularly, its generalized linear form can return to the standard linear Fréchet regression through a special choice of the weight function. Moreover, the generalized linear form possesses methodological and computational simplicity because the Euclidean variable and the metric space element are completely separable. The favorable theoretical properties (e.g. the estimation consistency and presentation theorem) of the nonlinear Fréchet regressions are established systemically. The comprehensive simulation studies and a human mortality data analysis demonstrate that the new strategy is significantly better than the competitors.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑