Issues in the estimation of mis-specified models of fractionally integrated processes

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This short paper provides a comprehensive set of new theoretical results on the impact of mis-specifying the short run dynamics in fractionally integrated processes. We show that four alternative parametric estimators – frequency domain maximum likelihood, Whittle, time domain maximum likelihood and conditional sum of squares – converge to the same pseudo-true value under common mis-specification, and that they possess a common asymptotic distribution. The results are derived assuming the true data generating mechanism is a fractional linear process driven by a martingale difference innovation. A completely general parametric specification for the short run dynamics of the estimated (mis-specified) fractional model is considered, and with long memory, short memory and antipersistence in both the model and the data generating mechanism accommodated. The paper can be seen as extending an existing line of research on mis-specification in fractional models, important contributions to which have appeared in Journal of Econometrics. It also complements a range of existing asymptotic results on estimation in correctly specified fractional models. Open problems in the area are the subject of the final discussion.

Original languageEnglish
Number of pages15
JournalJournal of Econometrics
DOIs
Publication statusAccepted/In press - 2019

Keywords

  • ARFIMA
  • Frequency domain estimators
  • Long memory models
  • Mis-specified short memory dynamics
  • Pseudo-true parameter
  • Time domain estimators

Cite this

@article{4fefa7c577f3428d8aa8d0fd7f6d6eb8,
title = "Issues in the estimation of mis-specified models of fractionally integrated processes",
abstract = "This short paper provides a comprehensive set of new theoretical results on the impact of mis-specifying the short run dynamics in fractionally integrated processes. We show that four alternative parametric estimators – frequency domain maximum likelihood, Whittle, time domain maximum likelihood and conditional sum of squares – converge to the same pseudo-true value under common mis-specification, and that they possess a common asymptotic distribution. The results are derived assuming the true data generating mechanism is a fractional linear process driven by a martingale difference innovation. A completely general parametric specification for the short run dynamics of the estimated (mis-specified) fractional model is considered, and with long memory, short memory and antipersistence in both the model and the data generating mechanism accommodated. The paper can be seen as extending an existing line of research on mis-specification in fractional models, important contributions to which have appeared in Journal of Econometrics. It also complements a range of existing asymptotic results on estimation in correctly specified fractional models. Open problems in the area are the subject of the final discussion.",
keywords = "ARFIMA, Frequency domain estimators, Long memory models, Mis-specified short memory dynamics, Pseudo-true parameter, Time domain estimators",
author = "Martin, {Gael M.} and K. Nadarajah and Poskitt, {D. S.}",
year = "2019",
doi = "10.1016/j.jeconom.2019.09.007",
language = "English",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier",

}

Issues in the estimation of mis-specified models of fractionally integrated processes. / Martin, Gael M.; Nadarajah, K.; Poskitt, D. S.

In: Journal of Econometrics, 2019.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Issues in the estimation of mis-specified models of fractionally integrated processes

AU - Martin, Gael M.

AU - Nadarajah, K.

AU - Poskitt, D. S.

PY - 2019

Y1 - 2019

N2 - This short paper provides a comprehensive set of new theoretical results on the impact of mis-specifying the short run dynamics in fractionally integrated processes. We show that four alternative parametric estimators – frequency domain maximum likelihood, Whittle, time domain maximum likelihood and conditional sum of squares – converge to the same pseudo-true value under common mis-specification, and that they possess a common asymptotic distribution. The results are derived assuming the true data generating mechanism is a fractional linear process driven by a martingale difference innovation. A completely general parametric specification for the short run dynamics of the estimated (mis-specified) fractional model is considered, and with long memory, short memory and antipersistence in both the model and the data generating mechanism accommodated. The paper can be seen as extending an existing line of research on mis-specification in fractional models, important contributions to which have appeared in Journal of Econometrics. It also complements a range of existing asymptotic results on estimation in correctly specified fractional models. Open problems in the area are the subject of the final discussion.

AB - This short paper provides a comprehensive set of new theoretical results on the impact of mis-specifying the short run dynamics in fractionally integrated processes. We show that four alternative parametric estimators – frequency domain maximum likelihood, Whittle, time domain maximum likelihood and conditional sum of squares – converge to the same pseudo-true value under common mis-specification, and that they possess a common asymptotic distribution. The results are derived assuming the true data generating mechanism is a fractional linear process driven by a martingale difference innovation. A completely general parametric specification for the short run dynamics of the estimated (mis-specified) fractional model is considered, and with long memory, short memory and antipersistence in both the model and the data generating mechanism accommodated. The paper can be seen as extending an existing line of research on mis-specification in fractional models, important contributions to which have appeared in Journal of Econometrics. It also complements a range of existing asymptotic results on estimation in correctly specified fractional models. Open problems in the area are the subject of the final discussion.

KW - ARFIMA

KW - Frequency domain estimators

KW - Long memory models

KW - Mis-specified short memory dynamics

KW - Pseudo-true parameter

KW - Time domain estimators

UR - http://www.scopus.com/inward/record.url?scp=85075436849&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2019.09.007

DO - 10.1016/j.jeconom.2019.09.007

M3 - Article

AN - SCOPUS:85075436849

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

ER -