Inference on the long-memory properties of time series with non-stationary volatility

Research output: Contribution to journal › Article › Research › peer review

Authors

  • Matei Demetrescu
  • Philipp Sibbertsen

Research Organisations

External Research Organisations

  • Kiel University
View graph of relations

Details

Original languageEnglish
Pages (from-to)80-84
Number of pages5
JournalEconomics letters
Volume144
Early online date4 May 2016
Publication statusPublished - Jul 2016

Abstract

Time-varying volatility is often present in time series data and can have adverse effects when inferring about the persistence properties of examined series. This note analyzes the effects of such nonstationarity on periodogram-based inference for the fractional integration parameter. Based on asymptotic arguments and Monte Carlo simulations, we show that the log-periodogram regression estimator remains consistent, but has asymptotic distribution whose variance depends on the variation of the volatility of the series.

Keywords

    Fractional integration, Heteroskedasticity, Modulated process, Persistence, Time-varying variance

ASJC Scopus subject areas

Cite this

Inference on the long-memory properties of time series with non-stationary volatility. / Demetrescu, Matei; Sibbertsen, Philipp.
In: Economics letters, Vol. 144, 07.2016, p. 80-84.

Research output: Contribution to journal › Article › Research › peer review

Demetrescu M, Sibbertsen P. Inference on the long-memory properties of time series with non-stationary volatility. Economics letters. 2016 Jul;144:80-84. Epub 2016 May 4. doi: 10.1016/j.econlet.2016.04.034
Demetrescu, Matei ; Sibbertsen, Philipp. / Inference on the long-memory properties of time series with non-stationary volatility. In: Economics letters. 2016 ; Vol. 144. pp. 80-84.
Download
@article{0ea1ed144f3f468c990dfc44445ccada,
title = "Inference on the long-memory properties of time series with non-stationary volatility",
abstract = "Time-varying volatility is often present in time series data and can have adverse effects when inferring about the persistence properties of examined series. This note analyzes the effects of such nonstationarity on periodogram-based inference for the fractional integration parameter. Based on asymptotic arguments and Monte Carlo simulations, we show that the log-periodogram regression estimator remains consistent, but has asymptotic distribution whose variance depends on the variation of the volatility of the series.",
keywords = "Fractional integration, Heteroskedasticity, Modulated process, Persistence, Time-varying variance",
author = "Matei Demetrescu and Philipp Sibbertsen",
note = "Funding Information: The authors would like to thank an anonymous referee, J{\"o}rg Breitung, Uwe Hassler, Liudas Giraitis and Maya Olivares for very helpful comments and suggestions, as well as Benjamin Hillmann for computational research assistance. The authors gratefully acknowledge the support of the Deutsche Forschungsgemeinschaft (DFG) through the projects DE 1617/4-1 and SI 745/9-1. ",
year = "2016",
month = jul,
doi = "10.1016/j.econlet.2016.04.034",
language = "English",
volume = "144",
pages = "80--84",
journal = "Economics letters",
issn = "0165-1765",
publisher = "Elsevier",

}

Download

TY - JOUR

T1 - Inference on the long-memory properties of time series with non-stationary volatility

AU - Demetrescu, Matei

AU - Sibbertsen, Philipp

N1 - Funding Information: The authors would like to thank an anonymous referee, Jörg Breitung, Uwe Hassler, Liudas Giraitis and Maya Olivares for very helpful comments and suggestions, as well as Benjamin Hillmann for computational research assistance. The authors gratefully acknowledge the support of the Deutsche Forschungsgemeinschaft (DFG) through the projects DE 1617/4-1 and SI 745/9-1.

PY - 2016/7

Y1 - 2016/7

N2 - Time-varying volatility is often present in time series data and can have adverse effects when inferring about the persistence properties of examined series. This note analyzes the effects of such nonstationarity on periodogram-based inference for the fractional integration parameter. Based on asymptotic arguments and Monte Carlo simulations, we show that the log-periodogram regression estimator remains consistent, but has asymptotic distribution whose variance depends on the variation of the volatility of the series.

AB - Time-varying volatility is often present in time series data and can have adverse effects when inferring about the persistence properties of examined series. This note analyzes the effects of such nonstationarity on periodogram-based inference for the fractional integration parameter. Based on asymptotic arguments and Monte Carlo simulations, we show that the log-periodogram regression estimator remains consistent, but has asymptotic distribution whose variance depends on the variation of the volatility of the series.

KW - Fractional integration

KW - Heteroskedasticity

KW - Modulated process

KW - Persistence

KW - Time-varying variance

UR - http://www.scopus.com/inward/record.url?scp=84967235620&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2016.04.034

DO - 10.1016/j.econlet.2016.04.034

M3 - Article

AN - SCOPUS:84967235620

VL - 144

SP - 80

EP - 84

JO - Economics letters

JF - Economics letters

SN - 0165-1765

ER -