Forecasting breaks and forecasting during breaks
Success in accurately forecasting breaks requires that they are predictable from relevant information available at the forecast origin using an appropriate model form, which can be selected and estimated before the break. To clarify the roles of these six necessary conditions, we distinguish betwee...
| Main Authors: | , , |
|---|---|
| Format: | Working paper |
| Published: |
University of Oxford
2011
|
| _version_ | 1826260774702546944 |
|---|---|
| author | Castle, J Hendry, D Fawcett, N |
| author_facet | Castle, J Hendry, D Fawcett, N |
| author_sort | Castle, J |
| collection | OXFORD |
| description | Success in accurately forecasting breaks requires that they are predictable from relevant information available at the forecast origin using an appropriate model form, which can be selected and estimated before the break. To clarify the roles of these six necessary conditions, we distinguish between the information set for 'normal forces' and the ones for 'break drivers', then outline sources of potential information. Relevant non-linear, dynamic models facing multiple breaks can have more candidate variables than observations, so we discuss automatic model selection. As a failure to accurately forecast breaks remains likely, we augment our strategy by modelling breaks during their progress, and consider robust forecasting devices. |
| first_indexed | 2024-03-06T19:11:03Z |
| format | Working paper |
| id | oxford-uuid:16c74267-6368-40f3-b84d-e7900d8aa976 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T19:11:03Z |
| publishDate | 2011 |
| publisher | University of Oxford |
| record_format | dspace |
| spelling | oxford-uuid:16c74267-6368-40f3-b84d-e7900d8aa9762022-03-26T10:33:20ZForecasting breaks and forecasting during breaksWorking paperhttp://purl.org/coar/resource_type/c_8042uuid:16c74267-6368-40f3-b84d-e7900d8aa976Bulk import via SwordSymplectic ElementsUniversity of Oxford2011Castle, JHendry, DFawcett, NSuccess in accurately forecasting breaks requires that they are predictable from relevant information available at the forecast origin using an appropriate model form, which can be selected and estimated before the break. To clarify the roles of these six necessary conditions, we distinguish between the information set for 'normal forces' and the ones for 'break drivers', then outline sources of potential information. Relevant non-linear, dynamic models facing multiple breaks can have more candidate variables than observations, so we discuss automatic model selection. As a failure to accurately forecast breaks remains likely, we augment our strategy by modelling breaks during their progress, and consider robust forecasting devices. |
| spellingShingle | Castle, J Hendry, D Fawcett, N Forecasting breaks and forecasting during breaks |
| title | Forecasting breaks and forecasting during breaks |
| title_full | Forecasting breaks and forecasting during breaks |
| title_fullStr | Forecasting breaks and forecasting during breaks |
| title_full_unstemmed | Forecasting breaks and forecasting during breaks |
| title_short | Forecasting breaks and forecasting during breaks |
| title_sort | forecasting breaks and forecasting during breaks |
| work_keys_str_mv | AT castlej forecastingbreaksandforecastingduringbreaks AT hendryd forecastingbreaksandforecastingduringbreaks AT fawcettn forecastingbreaksandforecastingduringbreaks |