Asymptotic theory for regressions with smoothly changing parameters

Eric Hillebrand; Marcelo  Medeiros; Junyue Xu

doi:10.1515/jtse-2012-0024

Asymptotic theory for regressions with smoothly changing parameters

Eric Hillebrand, Marcelo Medeiros, Junyue Xu

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avis › Tidsskriftartikel › Forskning › peer review

Abstract

We derive asymptotic properties of the quasi maximum likelihood estimator of
smooth transition regressions when time is the transition variable. The consistency of the estimator
and its asymptotic distribution are examined. It is shown that the estimator converges at
the usual pT-rate and has an asymptotically normal distribution. Finite sample properties of the
estimator are explored in simulations. We illustrate with an application to US inflation and output
data.

Originalsprog	Engelsk
Tidsskrift	Journal of Time Series Econometrics
Vol/bind	5
Nummer	2
Sider (fra-til)	133-162
Antal sider	33
ISSN	2194-6507
DOI	https://doi.org/10.1515/jtse-2012-0024
Status	Udgivet - 2013

Emneord

Regime switching
Smooth transition regression
Asymptotic theory

Adgang til dokumentet

10.1515/jtse-2012-0024

https://pure.au.dk/portal/files/45943708/rp12_31.pdf

Tjek tilgængelighed hos det Kgl. Bibliotek

Citationsformater

@article{86ffd5e2e97740acbdcba4ada785af9b,

title = "Asymptotic theory for regressions with smoothly changing parameters",

abstract = "We derive asymptotic properties of the quasi maximum likelihood estimator ofsmooth transition regressions when time is the transition variable. The consistency of the estimatorand its asymptotic distribution are examined. It is shown that the estimator converges atthe usual pT-rate and has an asymptotically normal distribution. Finite sample properties of theestimator are explored in simulations. We illustrate with an application to US inflation and outputdata.",

keywords = "Regime switching, Smooth transition regression, Asymptotic theory",

author = "Eric Hillebrand and Marcelo Medeiros and Junyue Xu",

year = "2013",

doi = "10.1515/jtse-2012-0024",

language = "English",

volume = "5",

pages = "133--162",

journal = "Journal of Time Series Econometrics",

issn = "2194-6507",

publisher = "Walter de Gruyter GmbH",

number = "2",

}

TY - JOUR

T1 - Asymptotic theory for regressions with smoothly changing parameters

AU - Hillebrand, Eric

AU - Medeiros, Marcelo

AU - Xu, Junyue

PY - 2013

Y1 - 2013

N2 - We derive asymptotic properties of the quasi maximum likelihood estimator ofsmooth transition regressions when time is the transition variable. The consistency of the estimatorand its asymptotic distribution are examined. It is shown that the estimator converges atthe usual pT-rate and has an asymptotically normal distribution. Finite sample properties of theestimator are explored in simulations. We illustrate with an application to US inflation and outputdata.

AB - We derive asymptotic properties of the quasi maximum likelihood estimator ofsmooth transition regressions when time is the transition variable. The consistency of the estimatorand its asymptotic distribution are examined. It is shown that the estimator converges atthe usual pT-rate and has an asymptotically normal distribution. Finite sample properties of theestimator are explored in simulations. We illustrate with an application to US inflation and outputdata.

KW - Regime switching

KW - Smooth transition regression

KW - Asymptotic theory

U2 - 10.1515/jtse-2012-0024

DO - 10.1515/jtse-2012-0024

M3 - Journal article

SN - 2194-6507

VL - 5

SP - 133

EP - 162

JO - Journal of Time Series Econometrics

JF - Journal of Time Series Econometrics

IS - 2

ER -