Evaluating automatic model selection
We evaluate automatically selecting the relevant variables in an econometric model from a large candidate set. General-to-specific selection is outlined for a constant model in orthogonal variables, where only one decision is required to select, irrespective of the number of regressors (N &...
| Main Authors: | , , |
|---|---|
| פורמט: | Working paper |
| יצא לאור: |
University of Oxford
2010
|
| _version_ | 1826272599363026944 |
|---|---|
| author | Castle, J Hendry, D Doornik, J |
| author_facet | Castle, J Hendry, D Doornik, J |
| author_sort | Castle, J |
| collection | OXFORD |
| description | We evaluate automatically selecting the relevant variables in an econometric model from a large candidate set. General-to-specific selection is outlined for a constant model in orthogonal variables, where only one decision is required to select, irrespective of the number of regressors (N < T) where T is the sample size, then evaluated in simulation experiments for N = 1000. Comparisons with Autometrics (Doornik, 2009) show similar properties, but not restricted to orthogonal cases. Monte Carlo experiments examine the roles of post-selection bias corrections and diagnostic testing, and evaluate Autometrics' capability in dynamic models by its cost of search versus costs of inference. |
| first_indexed | 2024-03-06T22:15:07Z |
| format | Working paper |
| id | oxford-uuid:53216ac6-b99e-43b2-8c8a-2f26b9c28079 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T22:15:07Z |
| publishDate | 2010 |
| publisher | University of Oxford |
| record_format | dspace |
| spelling | oxford-uuid:53216ac6-b99e-43b2-8c8a-2f26b9c280792022-03-26T16:29:41ZEvaluating automatic model selectionWorking paperhttp://purl.org/coar/resource_type/c_8042uuid:53216ac6-b99e-43b2-8c8a-2f26b9c28079Bulk import via SwordSymplectic ElementsUniversity of Oxford2010Castle, JHendry, DDoornik, JWe evaluate automatically selecting the relevant variables in an econometric model from a large candidate set. General-to-specific selection is outlined for a constant model in orthogonal variables, where only one decision is required to select, irrespective of the number of regressors (N < T) where T is the sample size, then evaluated in simulation experiments for N = 1000. Comparisons with Autometrics (Doornik, 2009) show similar properties, but not restricted to orthogonal cases. Monte Carlo experiments examine the roles of post-selection bias corrections and diagnostic testing, and evaluate Autometrics' capability in dynamic models by its cost of search versus costs of inference. |
| spellingShingle | Castle, J Hendry, D Doornik, J Evaluating automatic model selection |
| title | Evaluating automatic model selection |
| title_full | Evaluating automatic model selection |
| title_fullStr | Evaluating automatic model selection |
| title_full_unstemmed | Evaluating automatic model selection |
| title_short | Evaluating automatic model selection |
| title_sort | evaluating automatic model selection |
| work_keys_str_mv | AT castlej evaluatingautomaticmodelselection AT hendryd evaluatingautomaticmodelselection AT doornikj evaluatingautomaticmodelselection |