An automatic test of super exogeneity

We develop a new automatically-computable test for super exogeneity, using a variant of general-to-specific modeling. Based on the recent developments of impulse saturation applied to marginal models under the null that no impulses matter, we select the significant impulses for testing in the condi...

Ausführliche Beschreibung

Bibliographische Detailangaben
Hauptverfasser: Hendry, D, Santos, C
Format: Working paper
Veröffentlicht: University of Oxford 2010
_version_ 1826283562376101888
author Hendry, D
Santos, C
author_facet Hendry, D
Santos, C
author_sort Hendry, D
collection OXFORD
description We develop a new automatically-computable test for super exogeneity, using a variant of general-to-specific modeling. Based on the recent developments of impulse saturation applied to marginal models under the null that no impulses matter, we select the significant impulses for testing in the conditional. Since zero-mean changes are relatively undetectable in both VARs and conditional equations, we focus on location shifts, although we also discuss variance changes. The approximate analytical non-centrality of the test is derived for a failure of weak exogeneity when there is a shift in the marginal process. Monte Carlo simulations confirm the empirical accuracy of the nominal significance levels under the null, and show rejections for this failure of super exogeneity. An empirical application to UK M1 delivers new results for this much-studied data set.
first_indexed 2024-03-07T01:00:44Z
format Working paper
id oxford-uuid:899f89f9-52a2-4469-b7cb-5d25cb2d97c1
institution University of Oxford
last_indexed 2024-03-07T01:00:44Z
publishDate 2010
publisher University of Oxford
record_format dspace
spelling oxford-uuid:899f89f9-52a2-4469-b7cb-5d25cb2d97c12022-03-26T22:25:58ZAn automatic test of super exogeneityWorking paperhttp://purl.org/coar/resource_type/c_8042uuid:899f89f9-52a2-4469-b7cb-5d25cb2d97c1Bulk import via SwordSymplectic ElementsUniversity of Oxford2010Hendry, DSantos, CWe develop a new automatically-computable test for super exogeneity, using a variant of general-to-specific modeling. Based on the recent developments of impulse saturation applied to marginal models under the null that no impulses matter, we select the significant impulses for testing in the conditional. Since zero-mean changes are relatively undetectable in both VARs and conditional equations, we focus on location shifts, although we also discuss variance changes. The approximate analytical non-centrality of the test is derived for a failure of weak exogeneity when there is a shift in the marginal process. Monte Carlo simulations confirm the empirical accuracy of the nominal significance levels under the null, and show rejections for this failure of super exogeneity. An empirical application to UK M1 delivers new results for this much-studied data set.
spellingShingle Hendry, D
Santos, C
An automatic test of super exogeneity
title An automatic test of super exogeneity
title_full An automatic test of super exogeneity
title_fullStr An automatic test of super exogeneity
title_full_unstemmed An automatic test of super exogeneity
title_short An automatic test of super exogeneity
title_sort automatic test of super exogeneity
work_keys_str_mv AT hendryd anautomatictestofsuperexogeneity
AT santosc anautomatictestofsuperexogeneity
AT hendryd automatictestofsuperexogeneity
AT santosc automatictestofsuperexogeneity