Automatic selection for non-linear models
© 2012 Springer-Verlag London Limited.Our strategy for automatic selection in potentially non-linear processes is: test for non-linearity in the unrestricted linear formulation; if that test rejects, specify a general model using polynomials, to be simplified to a minimal congruent representation; f...
| Main Authors: | , |
|---|---|
| Formato: | Book section |
| Publicado: |
Springer London
2011
|
| _version_ | 1826298590962647040 |
|---|---|
| author | Castle, J Hendry, D |
| author_facet | Castle, J Hendry, D |
| author_sort | Castle, J |
| collection | OXFORD |
| description | © 2012 Springer-Verlag London Limited.Our strategy for automatic selection in potentially non-linear processes is: test for non-linearity in the unrestricted linear formulation; if that test rejects, specify a general model using polynomials, to be simplified to a minimal congruent representation; finally select by encompassing tests of specific non-linear forms against the selected model. Non-linearity poses many problems: extreme observations leading to non-normal (fat-tailed) distributions; collinearity between non-linear functions; usually more variables than observations when approximating the non-linearity; and excess retention of irrelevant variables; but solutions are proposed. A returns-to-education empirical application demonstrates the feasibility of the non-linear automatic model selection algorithm Autometrics. |
| first_indexed | 2024-03-07T04:49:12Z |
| format | Book section |
| id | oxford-uuid:d461083b-e451-483e-a7b3-81fbe2dbbbe3 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T04:49:12Z |
| publishDate | 2011 |
| publisher | Springer London |
| record_format | dspace |
| spelling | oxford-uuid:d461083b-e451-483e-a7b3-81fbe2dbbbe32022-03-27T08:18:02ZAutomatic selection for non-linear modelsBook sectionhttp://purl.org/coar/resource_type/c_3248uuid:d461083b-e451-483e-a7b3-81fbe2dbbbe3Symplectic Elements at OxfordSpringer London2011Castle, JHendry, D© 2012 Springer-Verlag London Limited.Our strategy for automatic selection in potentially non-linear processes is: test for non-linearity in the unrestricted linear formulation; if that test rejects, specify a general model using polynomials, to be simplified to a minimal congruent representation; finally select by encompassing tests of specific non-linear forms against the selected model. Non-linearity poses many problems: extreme observations leading to non-normal (fat-tailed) distributions; collinearity between non-linear functions; usually more variables than observations when approximating the non-linearity; and excess retention of irrelevant variables; but solutions are proposed. A returns-to-education empirical application demonstrates the feasibility of the non-linear automatic model selection algorithm Autometrics. |
| spellingShingle | Castle, J Hendry, D Automatic selection for non-linear models |
| title | Automatic selection for non-linear models |
| title_full | Automatic selection for non-linear models |
| title_fullStr | Automatic selection for non-linear models |
| title_full_unstemmed | Automatic selection for non-linear models |
| title_short | Automatic selection for non-linear models |
| title_sort | automatic selection for non linear models |
| work_keys_str_mv | AT castlej automaticselectionfornonlinearmodels AT hendryd automaticselectionfornonlinearmodels |