Wild Bootstrap Tests for Autocorrelation in Vector Autoregressive Models

Niklas Ahlgren, Paul Catani

Forskningsoutput: Kapitel i bok/rapport/konferenshandlingKonferensbidragVetenskapligPeer review

Sammanfattning

Tests for error autocorrelation (AC) are derived under the assumption of independent and identically distributed (IID) errors. The tests are not asymptotically valid if the errors are conditionally heteroskedastic.
In this paper we propose wild bootstrap (WB) Lagrange multiplier tests for error AC in vector autoregressive (VAR) models. We show that the WB tests are asymptotically valid under conditional heteroskedasticity of unknown form. WB tests based on a version of the heteroskedasticity-consistent covariance matrix estimator are found to have the smallest error in rejection probability under the null and high power under the alternative. We apply the tests to VAR models for credit default swap (CDS) prices and Euribor interest rates. An important result that we find is that the WB tests lead to parsimonious models while the asymptotic tests suggest that a long lag length is required to get white noise residuals.
OriginalspråkEngelska
Titel på gästpublikationProceedings of the 60th ISI World Statistics Congress : 26-31 July 2015, Rio de Janeiro, Brazil
Antal sidor6
UtgivningsortThe Hague
FörlagISI - International Statistical Institute
Utgivningsdatum12.2015
Sidor2259-2264
ISBN (elektroniskt)978-90-73592-35-3
StatusPublicerad - 12.2015
MoE-publikationstypA4 Artikel i en konferenspublikation
EvenemangISI World Statistics Congress (WSC) - Rio de Janeiro, Brasilien
Varaktighet: 26.07.201531.07.2015
Konferensnummer: 60

Fingeravtryck

Fördjupa i forskningsämnen för ”Wild Bootstrap Tests for Autocorrelation in Vector Autoregressive Models”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här