Nearly Efficient Likelihood Ratio Tests of a Unit Root in an Autoregressive Model of Arbitrary Order

Samuel Brien; Michael Jansson; Morten Ørregaard Nielsen

doi:10.1017/S0266466622000652

Nearly Efficient Likelihood Ratio Tests of a Unit Root in an Autoregressive Model of Arbitrary Order

Samuel Brien, Michael Jansson, Morten Ørregaard Nielsen

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Abstract

We study large sample properties of likelihood ratio tests of the unit-root hypothesis in an autoregressive model of arbitrary order. Earlier research on this testing problem has developed likelihood ratio tests in the autoregressive model of order 1, but resorted to a plug-in approach when dealing with higher-order models. In contrast, we consider the full model and derive the relevant large sample properties of likelihood ratio tests under a local-to-unity asymptotic framework. As in the simpler model, we show that the full likelihood ratio tests are nearly efficient, in the sense that their asymptotic local power functions are virtually indistinguishable from the Gaussian power envelopes. Extensions to sieve-type approximations and different classes of alternatives are also considered.

Original language	English
Journal	Econometric Theory
Volume	40
Issue	5
ISSN	0266-4666
DOIs	https://doi.org/10.1017/S0266466622000652
Publication status	Published - Oct 2024

Access to Document

10.1017/S0266466622000652Licence: CC BY

Cite this

@article{041f7555ae204c4d81b62a62af4b6228,

title = "Nearly Efficient Likelihood Ratio Tests of a Unit Root in an Autoregressive Model of Arbitrary Order",

abstract = "We study large sample properties of likelihood ratio tests of the unit-root hypothesis in an autoregressive model of arbitrary order. Earlier research on this testing problem has developed likelihood ratio tests in the autoregressive model of order 1, but resorted to a plug-in approach when dealing with higher-order models. In contrast, we consider the full model and derive the relevant large sample properties of likelihood ratio tests under a local-to-unity asymptotic framework. As in the simpler model, we show that the full likelihood ratio tests are nearly efficient, in the sense that their asymptotic local power functions are virtually indistinguishable from the Gaussian power envelopes. Extensions to sieve-type approximations and different classes of alternatives are also considered.",

author = "Samuel Brien and Michael Jansson and Nielsen, {Morten {\O}rregaard}",

year = "2024",

month = oct,

doi = "10.1017/S0266466622000652",

language = "English",

volume = "40",

journal = "Econometric Theory",

issn = "0266-4666",

publisher = "Cambridge University Press",

number = "5",

}

TY - JOUR

T1 - Nearly Efficient Likelihood Ratio Tests of a Unit Root in an Autoregressive Model of Arbitrary Order

AU - Brien, Samuel

AU - Jansson, Michael

AU - Nielsen, Morten Ørregaard

PY - 2024/10

Y1 - 2024/10

N2 - We study large sample properties of likelihood ratio tests of the unit-root hypothesis in an autoregressive model of arbitrary order. Earlier research on this testing problem has developed likelihood ratio tests in the autoregressive model of order 1, but resorted to a plug-in approach when dealing with higher-order models. In contrast, we consider the full model and derive the relevant large sample properties of likelihood ratio tests under a local-to-unity asymptotic framework. As in the simpler model, we show that the full likelihood ratio tests are nearly efficient, in the sense that their asymptotic local power functions are virtually indistinguishable from the Gaussian power envelopes. Extensions to sieve-type approximations and different classes of alternatives are also considered.

AB - We study large sample properties of likelihood ratio tests of the unit-root hypothesis in an autoregressive model of arbitrary order. Earlier research on this testing problem has developed likelihood ratio tests in the autoregressive model of order 1, but resorted to a plug-in approach when dealing with higher-order models. In contrast, we consider the full model and derive the relevant large sample properties of likelihood ratio tests under a local-to-unity asymptotic framework. As in the simpler model, we show that the full likelihood ratio tests are nearly efficient, in the sense that their asymptotic local power functions are virtually indistinguishable from the Gaussian power envelopes. Extensions to sieve-type approximations and different classes of alternatives are also considered.

UR - http://www.scopus.com/inward/record.url?scp=85144917634&partnerID=8YFLogxK

U2 - 10.1017/S0266466622000652

DO - 10.1017/S0266466622000652

M3 - Journal article

SN - 0266-4666

VL - 40

JO - Econometric Theory

JF - Econometric Theory

IS - 5

ER -

Nearly Efficient Likelihood Ratio Tests of a Unit Root in an Autoregressive Model of Arbitrary Order

Abstract

Access to Document

Other files and links

Fingerprint

Cite this