A note on estimation in Hilbertian linear models

Siegfried Hörmann, Lukasz Kidziski

Research output: Contribution to journalArticle

Abstract

We study estimation and prediction in linear models where the response and the regressor variable both take values in some Hilbert space. Our main objective is to obtain consistency of a principal component-based estimator for the regression operator under minimal assumptions. In particular, we avoid some inconvenient technical restrictions that have been used throughout the literature. We develop our theory in a time-dependent setup that comprises as important special case the autoregressive Hilbertian model.
LanguageUndefined/Unknown
Pages43-62
Number of pages20
JournalScandinavian journal of statistics
Volume42
Issue number1
StatusPublished - 2015

Cite this

A note on estimation in Hilbertian linear models. / Hörmann, Siegfried; Kidziski, Lukasz.

In: Scandinavian journal of statistics, Vol. 42, No. 1, 2015, p. 43-62.

Research output: Contribution to journalArticle

@article{c24d1324c85e405ea2633583c890e426,
title = "A note on estimation in Hilbertian linear models",
abstract = "We study estimation and prediction in linear models where the response and the regressor variable both take values in some Hilbert space. Our main objective is to obtain consistency of a principal component-based estimator for the regression operator under minimal assumptions. In particular, we avoid some inconvenient technical restrictions that have been used throughout the literature. We develop our theory in a time-dependent setup that comprises as important special case the autoregressive Hilbertian model.",
author = "Siegfried H{\"o}rmann and Lukasz Kidziski",
note = "DOI: 10.1111/sjos.12094",
year = "2015",
language = "undefiniert/unbekannt",
volume = "42",
pages = "43--62",
journal = "Scandinavian journal of statistics",
issn = "0303-6898",
publisher = "Wiley-Blackwell",
number = "1",

}

TY - JOUR

T1 - A note on estimation in Hilbertian linear models

AU - Hörmann,Siegfried

AU - Kidziski,Lukasz

N1 - DOI: 10.1111/sjos.12094

PY - 2015

Y1 - 2015

N2 - We study estimation and prediction in linear models where the response and the regressor variable both take values in some Hilbert space. Our main objective is to obtain consistency of a principal component-based estimator for the regression operator under minimal assumptions. In particular, we avoid some inconvenient technical restrictions that have been used throughout the literature. We develop our theory in a time-dependent setup that comprises as important special case the autoregressive Hilbertian model.

AB - We study estimation and prediction in linear models where the response and the regressor variable both take values in some Hilbert space. Our main objective is to obtain consistency of a principal component-based estimator for the regression operator under minimal assumptions. In particular, we avoid some inconvenient technical restrictions that have been used throughout the literature. We develop our theory in a time-dependent setup that comprises as important special case the autoregressive Hilbertian model.

M3 - Artikel

VL - 42

SP - 43

EP - 62

JO - Scandinavian journal of statistics

T2 - Scandinavian journal of statistics

JF - Scandinavian journal of statistics

SN - 0303-6898

IS - 1

ER -