Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks
We present a novel Bayesian nonparametric regression model for covariates X and continuous response variable Y∈R. The model is parametrized in terms of marginal distributions for Y and X and a regression function which tunes the stochastic ordering of the conditional distributions F(y|x). By adoptin...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Institute of Mathematical Statistics
2016
|
| _version_ | 1826258712185012224 |
|---|---|
| author | Gray-Davies, T Holmes, C Caron, F |
| author_facet | Gray-Davies, T Holmes, C Caron, F |
| author_sort | Gray-Davies, T |
| collection | OXFORD |
| description | We present a novel Bayesian nonparametric regression model for covariates X and continuous response variable Y∈R. The model is parametrized in terms of marginal distributions for Y and X and a regression function which tunes the stochastic ordering of the conditional distributions F(y|x). By adopting an approximate composite likelihood approach, we show that the resulting posterior inference can be decoupled for the separate components of the model. This procedure can scale to very large datasets and allows for the use of standard, existing, software from Bayesian nonparametric density estimation and Plackett-Luce ranking estimation to be applied. As an illustration, we show an application of our approach to a US Census dataset, with over 1,300,000 data points and more than 100 covariates. |
| first_indexed | 2024-03-06T18:38:18Z |
| format | Journal article |
| id | oxford-uuid:0c0c185c-7d5d-430e-a6c1-a46a92007e56 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T18:38:18Z |
| publishDate | 2016 |
| publisher | Institute of Mathematical Statistics |
| record_format | dspace |
| spelling | oxford-uuid:0c0c185c-7d5d-430e-a6c1-a46a92007e562022-03-26T09:32:44ZScalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranksJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:0c0c185c-7d5d-430e-a6c1-a46a92007e56EnglishSymplectic Elements at OxfordInstitute of Mathematical Statistics2016Gray-Davies, THolmes, CCaron, FWe present a novel Bayesian nonparametric regression model for covariates X and continuous response variable Y∈R. The model is parametrized in terms of marginal distributions for Y and X and a regression function which tunes the stochastic ordering of the conditional distributions F(y|x). By adopting an approximate composite likelihood approach, we show that the resulting posterior inference can be decoupled for the separate components of the model. This procedure can scale to very large datasets and allows for the use of standard, existing, software from Bayesian nonparametric density estimation and Plackett-Luce ranking estimation to be applied. As an illustration, we show an application of our approach to a US Census dataset, with over 1,300,000 data points and more than 100 covariates. |
| spellingShingle | Gray-Davies, T Holmes, C Caron, F Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title | Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title_full | Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title_fullStr | Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title_full_unstemmed | Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title_short | Scalable Bayesian nonparametric regression via a Plackett-Luce model for conditional ranks |
| title_sort | scalable bayesian nonparametric regression via a plackett luce model for conditional ranks |
| work_keys_str_mv | AT graydaviest scalablebayesiannonparametricregressionviaaplackettlucemodelforconditionalranks AT holmesc scalablebayesiannonparametricregressionviaaplackettlucemodelforconditionalranks AT caronf scalablebayesiannonparametricregressionviaaplackettlucemodelforconditionalranks |