Model Identification and Non-unique Structure.
Identification is an essential attribute of any model's parameters, so we consider its three aspects of 'uniqueness', 'correspondence to reality' and 'interpretability'. Observationally-equivalent over-identified models can co-exist, and are mutually encompassing i...
Main Authors: | , , |
---|---|
Formato: | Working paper |
Idioma: | English |
Publicado em: |
Nuffield College (University of Oxford)
2001
|
_version_ | 1826260894983651328 |
---|---|
author | Hendry, D Lu, M Mizon, G |
author_facet | Hendry, D Lu, M Mizon, G |
author_sort | Hendry, D |
collection | OXFORD |
description | Identification is an essential attribute of any model's parameters, so we consider its three aspects of 'uniqueness', 'correspondence to reality' and 'interpretability'. Observationally-equivalent over-identified models can co-exist, and are mutually encompassing in the population; correctly-identified models need not correspond to the underlying structure; and may be wrongly interpreted. That a given model is over-identified with all over-identifying restrictions valid (even asymptotically) is insufficient to demonstrate that it is a unique representation. Moreover, structre (as invariance under extended information) need not be identifiable. We consider the role of structural breaks to discriminate between such representations. |
first_indexed | 2024-03-06T19:12:58Z |
format | Working paper |
id | oxford-uuid:17627e20-797f-4cc6-97ef-9733cc30a84c |
institution | University of Oxford |
language | English |
last_indexed | 2024-03-06T19:12:58Z |
publishDate | 2001 |
publisher | Nuffield College (University of Oxford) |
record_format | dspace |
spelling | oxford-uuid:17627e20-797f-4cc6-97ef-9733cc30a84c2022-03-26T10:36:59ZModel Identification and Non-unique Structure.Working paperhttp://purl.org/coar/resource_type/c_8042uuid:17627e20-797f-4cc6-97ef-9733cc30a84cEnglishDepartment of Economics - ePrintsNuffield College (University of Oxford)2001Hendry, DLu, MMizon, GIdentification is an essential attribute of any model's parameters, so we consider its three aspects of 'uniqueness', 'correspondence to reality' and 'interpretability'. Observationally-equivalent over-identified models can co-exist, and are mutually encompassing in the population; correctly-identified models need not correspond to the underlying structure; and may be wrongly interpreted. That a given model is over-identified with all over-identifying restrictions valid (even asymptotically) is insufficient to demonstrate that it is a unique representation. Moreover, structre (as invariance under extended information) need not be identifiable. We consider the role of structural breaks to discriminate between such representations. |
spellingShingle | Hendry, D Lu, M Mizon, G Model Identification and Non-unique Structure. |
title | Model Identification and Non-unique Structure. |
title_full | Model Identification and Non-unique Structure. |
title_fullStr | Model Identification and Non-unique Structure. |
title_full_unstemmed | Model Identification and Non-unique Structure. |
title_short | Model Identification and Non-unique Structure. |
title_sort | model identification and non unique structure |
work_keys_str_mv | AT hendryd modelidentificationandnonuniquestructure AT lum modelidentificationandnonuniquestructure AT mizong modelidentificationandnonuniquestructure |