R-squared for Bayesian regression models

dc.contributorAalto-yliopistofi
dc.contributorAalto Universityen
dc.contributor.authorGelman, Andrewen_US
dc.contributor.authorGoodrich, Benen_US
dc.contributor.authorGabry, Jonahen_US
dc.contributor.authorVehtari, Akien_US
dc.contributor.departmentDepartment of Computer Scienceen
dc.contributor.groupauthorProfessorship Vehtari Akien
dc.contributor.groupauthorHelsinki Institute for Information Technology (HIIT)en
dc.contributor.groupauthorProbabilistic Machine Learningen
dc.contributor.organizationColumbia Universityen_US
dc.date.accessioned2019-06-20T13:16:43Z
dc.date.available2019-06-20T13:16:43Z
dc.date.embargoinfo:eu-repo/date/embargoEnd/2020-05-13en_US
dc.date.issued2019-01-01en_US
dc.description.abstractThe usual definition of R 2 (variance of the predicted values divided by the variance of the data) has a problem for Bayesian fits, as the numerator can be larger than the denominator. We propose an alternative definition similar to one that has appeared in the survival analysis literature: the variance of the predicted values divided by the variance of predicted values plus the expected variance of the errors.en
dc.description.versionPeer revieweden
dc.format.extent6
dc.format.mimetypeapplication/pdfen_US
dc.identifier.citationGelman, A, Goodrich, B, Gabry, J & Vehtari, A 2019, 'R-squared for Bayesian regression models', American Statistician, pp. 1-6. https://doi.org/10.1080/00031305.2018.1549100en
dc.identifier.doi10.1080/00031305.2018.1549100en_US
dc.identifier.issn0003-1305
dc.identifier.issn1537-2731
dc.identifier.otherPURE UUID: d31122a0-cec7-477e-a87e-f6ec2e6ab8aeen_US
dc.identifier.otherPURE ITEMURL: https://research.aalto.fi/en/publications/d31122a0-cec7-477e-a87e-f6ec2e6ab8aeen_US
dc.identifier.otherPURE LINK: https://doi.org/10.1080/00031305.2018.1549100
dc.identifier.otherPURE FILEURL: https://research.aalto.fi/files/34206843/bayes_R2_v3.pdf
dc.identifier.urihttps://aaltodoc.aalto.fi/handle/123456789/38878
dc.identifier.urnURN:NBN:fi:aalto-201906203944
dc.language.isoenen
dc.publisherTaylor & Francis
dc.relation.ispartofseriesAmerican Statisticianen
dc.relation.ispartofseriespp. 1-6en
dc.rightsopenAccessen
dc.subject.keywordBayesian methodsen_US
dc.subject.keywordR-squareden_US
dc.subject.keywordRegressionen_US
dc.titleR-squared for Bayesian regression modelsen
dc.typeA1 Alkuperäisartikkeli tieteellisessä aikakauslehdessäfi
dc.type.versionacceptedVersion

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
bayes_R2_v3.pdf
Size:
469.81 KB
Format:
Adobe Portable Document Format