arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2001.11168stat.MLcs.LGmath.STstat.COstat.TH
收藏

模态线性回归的核选择:最优核与IRLS算法

Kernel Selection for Modal Linear Regression: Optimal Kernel and IRLS Algorithm

Kyoto University(京都大学)

Ryoya Yamasaki, Toshiyuki Tanaka

摘要:

模态线性回归(MLR)是用于获得作为线性模型的条件模式预测器的方法。我们从两个角度研究了MLR的核选择:“哪一个核实现更小的误差?和“哪个内核在计算上是有效的?".首先,我们证明了一个双权内核是最佳的意义上最小化的渐近均方误差的MLR参数。这一结果来自我们对MLR的渐近统计行为的精细分析。其次,我们提供了一个核类,迭代加权最小二乘算法(IRLS)是保证收敛的,特别是证明了IRLS与Epanechnikov核终止于有限次迭代。仿真研究经验验证,使用双权内核提供了良好的估计精度,使用Epanechnikov内核是计算效率。我们的结果改善了MLR,现有的研究往往坚持高斯核和模态期望最大化算法专门为它,通过提供核选择的指导方针。

英文摘要:

Modal linear regression (MLR) is a method for obtaining a conditional mode predictor as a linear model. We study kernel selection for MLR from two perspectives: "which kernel achieves smaller error?" and "which kernel is computationally efficient?". First, we show that a Biweight kernel is optimal in the sense of minimizing an asymptotic mean squared error of a resulting MLR parameter. This result is derived from our refined analysis of an asymptotic statistical behavior of MLR. Secondly, we provide a kernel class for which iteratively reweighted least-squares algorithm (IRLS) is guaranteed to converge, and especially prove that IRLS with an Epanechnikov kernel terminates in a finite number of iterations. Simulation studies empirically verified that using a Biweight kernel provides good estimation accuracy and that using an Epanechnikov kernel is computationally efficient. Our results improve MLR of which existing studies often stick to a Gaussian kernel and modal EM algorithm specialized for it, by providing guidelines of kernel selection.

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。

↑