Estimation of the noise covariance operator in functional linear regression with functional outputs

被引:2
作者
Crambes, Christophe [1 ]
Hilgert, Nadine [2 ]
Manrique, Tito [1 ,2 ]
机构
[1] Univ Montpellier, UMR IMAG 5149, Pl Eugene Bataillon, F-34095 Montpellier 5, France
[2] INRA, UMR MISTEA 0729, 2 Pl Pierre Viala, F-34060 Montpellier 2, France
关键词
Functional linear regression; Functional response; Noise covariance operator;
D O I
10.1016/j.spl.2016.02.006
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
This work deals with the estimation of the noise in functional linear regression when both the response and the covariate are functional. Namely, we propose two estimators of the covariance operator of the noise. We give some asymptotic properties of these estimators, and we study their behavior on simulations. (C) 2016 Elsevier B.V. All rights reserved.
引用
收藏
页码:7 / 15
页数:9
相关论文
共 8 条
[1]  
Ash R. B., 1975, PROBABILITY MATH STA, V27
[2]   Functional linear model [J].
Cardot, H ;
Ferraty, F ;
Sarda, P .
STATISTICS & PROBABILITY LETTERS, 1999, 45 (01) :11-22
[3]   Asymptotics of prediction in functional linear regression with functional outputs [J].
Crambes, Christophe ;
Mas, Andre .
BERNOULLI, 2013, 19 (5B) :2627-2651
[4]  
Horvath L., 2012, Inference for functional data with applications
[5]  
Hsing T., 2015, THEORETICAL FDN FUNC
[6]  
James R., 2005, Functional data analysis, V2nd
[7]  
Johnson R. A, 2007, APPL MULTIVARIATE ST, V6th, DOI DOI 10.4236/JWARP.2010.26066
[8]  
Wahba G., 1991, SPLINE MODELS OBSERV