Nonparametric estimation of a latent variable model

被引:8
|
作者
Kelava, Augustin [1 ]
Kohler, Michael [2 ]
Krzyzak, Adam [3 ]
Schaffland, Tim Fabian [1 ]
机构
[1] Univ Tubingen, Hector Inst Empir Bldg Forsch, Wirtschafts & Sozialwissensch Fak, Europastr 6, D-72072 Tubingen, Germany
[2] Tech Univ Darmstadt, Fachbereich Math, Schlossgartenstr 7, D-64289 Darmstadt, Germany
[3] Concordia Univ, Dept Comp Sci & Software Engn, 1455 Boul Maisonneuve Ouest, Montreal, PQ H3G 1M8, Canada
基金
加拿大自然科学与工程研究理事会;
关键词
Common factor analysis; Latent variables; Nonparametric regression; Consistency; SEMIPARAMETRIC APPROACH; NONLINEAR RELATIONS; REGRESSION; CONVERGENCE; CONSISTENCY; SELECTION; THEOREM; ERROR; LMS;
D O I
10.1016/j.jmva.2016.10.006
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
In this paper a nonparametric latent variable model is estimated without specifying the underlying distributions. The main idea is to estimate in a first step a common factor analysis model under the assumption that each manifest variable is influenced by at most one of the latent variables. In a second step nonparametric regression is used to analyze the relation between the latent variables. Theoretical results concerning consistency of the estimates are presented, and the finite sample size performance of the estimates is illustrated by applying them to simulated data. (C) 2016 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
引用
收藏
页码:112 / 134
页数:23
相关论文
共 50 条