Asymptotics of least squares for nonlinear harmonic regression

被引:1
|
作者
Song, Kai-Sheng [1 ]
Li, Ta-Hsin [2 ]
机构
[1] Univ N Texas, Dept Math, Denton, TX 76203 USA
[2] IBM TJ Watson Res Ctr, Dept Math Sci, Yorktown Hts, NY 10598 USA
关键词
frequency; harmonic retrieval; hidden periodicity; mixed spectrum; nonlinear least squares; signal processing; TIME-SERIES; MAXIMUM; ESTIMATOR;
D O I
10.1080/02331881003797593
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
This paper considers the least-squares estimator of a nonlinear regression problem where the regression function is a sum of sinusoids with unknown amplitudes and frequencies. It provides a complete proof of the consistency and the asymptotic normality of the nonlinear least-squares estimator after an expository discussion on the gaps and flaws in the related literature.
引用
收藏
页码:309 / 318
页数:10
相关论文
共 50 条