Preliminary test estimation for regression models with long-memory disturbance

Masanobu Taniguchi, Hiroaki Ogata, Hiroshi Shiraishi

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

For a class of time series regression models with long-memory disturbance, we are interested in estimation of a subset of the regression coefficient vector and spectral parameter of the residual process when the complementary subset is suspected to be close to 0. In this situation, we evaluate the mean square errors of the restricted and unrestricted MLE and a preliminary test estimator when the complementary parameters are contiguous to zero vector. The results are expressed in terms of the regression spectra and the residual spectra. Since we assume long-memory dependence for the disturbance, the asymptotics are much different from the case of i.i.d. disturbance. Numerical studies elucidate some interesting features of regression and long-memory structures.

Original languageEnglish
Pages (from-to)3213-3224
Number of pages12
JournalCommunications in Statistics - Theory and Methods
Volume38
Issue number16-17
DOIs
Publication statusPublished - 2009 Jan
Externally publishedYes

Fingerprint

Preliminary Test
Long Memory
Regression Model
Disturbance
Regression
Preliminary Test Estimator
Zero vector
Subset
Time Series Models
Regression Coefficient
Mean square error
Numerical Study
Evaluate

Keywords

  • Fractional spectral density
  • LAN theorem
  • Long-memory process
  • Preliminary test estimator
  • Restricted MLE
  • Time regression model
  • Unrestricted MLE

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Preliminary test estimation for regression models with long-memory disturbance. / Taniguchi, Masanobu; Ogata, Hiroaki; Shiraishi, Hiroshi.

In: Communications in Statistics - Theory and Methods, Vol. 38, No. 16-17, 01.2009, p. 3213-3224.

Research output: Contribution to journalArticle

@article{6fe9502880334572b67ee7a8b8e233fd,
title = "Preliminary test estimation for regression models with long-memory disturbance",
abstract = "For a class of time series regression models with long-memory disturbance, we are interested in estimation of a subset of the regression coefficient vector and spectral parameter of the residual process when the complementary subset is suspected to be close to 0. In this situation, we evaluate the mean square errors of the restricted and unrestricted MLE and a preliminary test estimator when the complementary parameters are contiguous to zero vector. The results are expressed in terms of the regression spectra and the residual spectra. Since we assume long-memory dependence for the disturbance, the asymptotics are much different from the case of i.i.d. disturbance. Numerical studies elucidate some interesting features of regression and long-memory structures.",
keywords = "Fractional spectral density, LAN theorem, Long-memory process, Preliminary test estimator, Restricted MLE, Time regression model, Unrestricted MLE",
author = "Masanobu Taniguchi and Hiroaki Ogata and Hiroshi Shiraishi",
year = "2009",
month = "1",
doi = "10.1080/03610920902947741",
language = "English",
volume = "38",
pages = "3213--3224",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "16-17",

}

TY - JOUR

T1 - Preliminary test estimation for regression models with long-memory disturbance

AU - Taniguchi, Masanobu

AU - Ogata, Hiroaki

AU - Shiraishi, Hiroshi

PY - 2009/1

Y1 - 2009/1

N2 - For a class of time series regression models with long-memory disturbance, we are interested in estimation of a subset of the regression coefficient vector and spectral parameter of the residual process when the complementary subset is suspected to be close to 0. In this situation, we evaluate the mean square errors of the restricted and unrestricted MLE and a preliminary test estimator when the complementary parameters are contiguous to zero vector. The results are expressed in terms of the regression spectra and the residual spectra. Since we assume long-memory dependence for the disturbance, the asymptotics are much different from the case of i.i.d. disturbance. Numerical studies elucidate some interesting features of regression and long-memory structures.

AB - For a class of time series regression models with long-memory disturbance, we are interested in estimation of a subset of the regression coefficient vector and spectral parameter of the residual process when the complementary subset is suspected to be close to 0. In this situation, we evaluate the mean square errors of the restricted and unrestricted MLE and a preliminary test estimator when the complementary parameters are contiguous to zero vector. The results are expressed in terms of the regression spectra and the residual spectra. Since we assume long-memory dependence for the disturbance, the asymptotics are much different from the case of i.i.d. disturbance. Numerical studies elucidate some interesting features of regression and long-memory structures.

KW - Fractional spectral density

KW - LAN theorem

KW - Long-memory process

KW - Preliminary test estimator

KW - Restricted MLE

KW - Time regression model

KW - Unrestricted MLE

UR - http://www.scopus.com/inward/record.url?scp=70249143971&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70249143971&partnerID=8YFLogxK

U2 - 10.1080/03610920902947741

DO - 10.1080/03610920902947741

M3 - Article

AN - SCOPUS:70249143971

VL - 38

SP - 3213

EP - 3224

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 16-17

ER -