qgam: Bayesian Nonparametric Quantile Regression Modeling in R
Generalized additive models (GAMs) are flexible non-linear regression models, which can be fitted efficiently using the approximate Bayesian methods provided by the mgcv R package. While the GAM methods provided by mgcv are based on the assumption that the response distribution is modeled parametric...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Foundation for Open Access Statistics
2021-11-01
|
| Series: | Journal of Statistical Software |
| Subjects: | |
| Online Access: | https://www.jstatsoft.org/index.php/jss/article/view/3800 |
| _version_ | 1797813956937515008 |
|---|---|
| author | Matteo Fasiolo Simon N. Wood Margaux Zaffran Raphaël Nedellec Yannig Goude |
| author_facet | Matteo Fasiolo Simon N. Wood Margaux Zaffran Raphaël Nedellec Yannig Goude |
| author_sort | Matteo Fasiolo |
| collection | DOAJ |
| description | Generalized additive models (GAMs) are flexible non-linear regression models, which can be fitted efficiently using the approximate Bayesian methods provided by the mgcv R package. While the GAM methods provided by mgcv are based on the assumption that the response distribution is modeled parametrically, here we discuss more flexible methods that do not entail any parametric assumption. In particular, this article introduces the qgam package, which is an extension of mgcv providing fast calibrated Bayesian methods for fitting quantile GAMs (QGAMs) in R. QGAMs are based on a smooth version of the pinball loss of Koenker (2005), rather than on a likelihood function, hence jointly achieving satisfactory accuracy of the quantile point estimates and coverage of the corresponding credible intervals requires adopting the specialized Bayesian fitting framework of Fasiolo, Wood, Zaffran, Nedellec, and Goude (2021b). Here we detail how this framework is implemented in qgam and we provide examples illustrating how the package should be used in practice. |
| first_indexed | 2024-03-13T08:00:23Z |
| format | Article |
| id | doaj.art-b2f599d50a774eab989c491c66b5af79 |
| institution | Directory Open Access Journal |
| issn | 1548-7660 |
| language | English |
| last_indexed | 2024-03-13T08:00:23Z |
| publishDate | 2021-11-01 |
| publisher | Foundation for Open Access Statistics |
| record_format | Article |
| series | Journal of Statistical Software |
| spelling | doaj.art-b2f599d50a774eab989c491c66b5af792023-06-01T18:48:04ZengFoundation for Open Access StatisticsJournal of Statistical Software1548-76602021-11-0110013110.18637/jss.v100.i093619qgam: Bayesian Nonparametric Quantile Regression Modeling in RMatteo Fasiolo0https://orcid.org/0000-0003-2335-5536Simon N. Wood1https://orcid.org/0000-0002-2034-7453Margaux Zaffran2https://orcid.org/0000-0001-7560-8916Raphaël Nedellec3https://orcid.org/0000-0001-7714-5177Yannig Goude4https://orcid.org/0000-0003-2028-5536University of BristolUniversity of BristolENSTA ParisÉlectricité de France R&DÉlectricité de France R&DGeneralized additive models (GAMs) are flexible non-linear regression models, which can be fitted efficiently using the approximate Bayesian methods provided by the mgcv R package. While the GAM methods provided by mgcv are based on the assumption that the response distribution is modeled parametrically, here we discuss more flexible methods that do not entail any parametric assumption. In particular, this article introduces the qgam package, which is an extension of mgcv providing fast calibrated Bayesian methods for fitting quantile GAMs (QGAMs) in R. QGAMs are based on a smooth version of the pinball loss of Koenker (2005), rather than on a likelihood function, hence jointly achieving satisfactory accuracy of the quantile point estimates and coverage of the corresponding credible intervals requires adopting the specialized Bayesian fitting framework of Fasiolo, Wood, Zaffran, Nedellec, and Goude (2021b). Here we detail how this framework is implemented in qgam and we provide examples illustrating how the package should be used in practice.https://www.jstatsoft.org/index.php/jss/article/view/3800bayesian quantile regressiongeneralized additive modelsregression splinescalibrated bayesfast bayesian inferencer |
| spellingShingle | Matteo Fasiolo Simon N. Wood Margaux Zaffran Raphaël Nedellec Yannig Goude qgam: Bayesian Nonparametric Quantile Regression Modeling in R Journal of Statistical Software bayesian quantile regression generalized additive models regression splines calibrated bayes fast bayesian inference r |
| title | qgam: Bayesian Nonparametric Quantile Regression Modeling in R |
| title_full | qgam: Bayesian Nonparametric Quantile Regression Modeling in R |
| title_fullStr | qgam: Bayesian Nonparametric Quantile Regression Modeling in R |
| title_full_unstemmed | qgam: Bayesian Nonparametric Quantile Regression Modeling in R |
| title_short | qgam: Bayesian Nonparametric Quantile Regression Modeling in R |
| title_sort | qgam bayesian nonparametric quantile regression modeling in r |
| topic | bayesian quantile regression generalized additive models regression splines calibrated bayes fast bayesian inference r |
| url | https://www.jstatsoft.org/index.php/jss/article/view/3800 |
| work_keys_str_mv | AT matteofasiolo qgambayesiannonparametricquantileregressionmodelinginr AT simonnwood qgambayesiannonparametricquantileregressionmodelinginr AT margauxzaffran qgambayesiannonparametricquantileregressionmodelinginr AT raphaelnedellec qgambayesiannonparametricquantileregressionmodelinginr AT yanniggoude qgambayesiannonparametricquantileregressionmodelinginr |