Evaluation of methods for interval estimation of model outputs, with application to survival models
When a published statistical model is also distributed as computer software, it will usually be desirable to present the outputs as interval, as well as point, estimates. The present paper compares three methods for approximate interval estimation about a model output, for use when the model form do...
| Main Author: | |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
2003
|
| _version_ | 1826294307799171072 |
|---|---|
| author | Stevens, R |
| author_facet | Stevens, R |
| author_sort | Stevens, R |
| collection | OXFORD |
| description | When a published statistical model is also distributed as computer software, it will usually be desirable to present the outputs as interval, as well as point, estimates. The present paper compares three methods for approximate interval estimation about a model output, for use when the model form does not permit an exact interval estimate. The methods considered are first-order asymptotics, using second derivatives of the log-likelihood to estimate variance information; higher-order asymptotics based on the signed-root transformation; and the non-parametric bootstrap. The signed-root method is Bayesian, and uses an approximation for posterior moments that has not previously been tested in a real-world application. Use of the three methods is illustrated with reference to a software project arising in medical decision-making, the UKPDS Risk Engine. Intervals from the first-order and signed-root methods are near-identical, and typically 1% wider to 7% narrower than those from the non-parametric bootstrap. The asymptotic methods are markedly faster than the bootstrap method. |
| first_indexed | 2024-03-07T03:43:37Z |
| format | Journal article |
| id | oxford-uuid:beb6585e-253b-4e6a-99d5-12089ee9a38f |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T03:43:37Z |
| publishDate | 2003 |
| record_format | dspace |
| spelling | oxford-uuid:beb6585e-253b-4e6a-99d5-12089ee9a38f2022-03-27T05:41:47ZEvaluation of methods for interval estimation of model outputs, with application to survival modelsJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:beb6585e-253b-4e6a-99d5-12089ee9a38fEnglishSymplectic Elements at Oxford2003Stevens, RWhen a published statistical model is also distributed as computer software, it will usually be desirable to present the outputs as interval, as well as point, estimates. The present paper compares three methods for approximate interval estimation about a model output, for use when the model form does not permit an exact interval estimate. The methods considered are first-order asymptotics, using second derivatives of the log-likelihood to estimate variance information; higher-order asymptotics based on the signed-root transformation; and the non-parametric bootstrap. The signed-root method is Bayesian, and uses an approximation for posterior moments that has not previously been tested in a real-world application. Use of the three methods is illustrated with reference to a software project arising in medical decision-making, the UKPDS Risk Engine. Intervals from the first-order and signed-root methods are near-identical, and typically 1% wider to 7% narrower than those from the non-parametric bootstrap. The asymptotic methods are markedly faster than the bootstrap method. |
| spellingShingle | Stevens, R Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title | Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title_full | Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title_fullStr | Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title_full_unstemmed | Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title_short | Evaluation of methods for interval estimation of model outputs, with application to survival models |
| title_sort | evaluation of methods for interval estimation of model outputs with application to survival models |
| work_keys_str_mv | AT stevensr evaluationofmethodsforintervalestimationofmodeloutputswithapplicationtosurvivalmodels |