deepregression: A Flexible Neural Network Framework for Semi-Structured Deep Distributional Regression
dc.contributor.author
Rügamer, David
dc.contributor.author
Kolb, Chris
dc.contributor.author
Fritz, Cornelius
dc.contributor.author
Pfisterer, Florian
dc.contributor.author
Kopper, Philipp
dc.contributor.author
Bischl, Bernd
dc.contributor.author
Shen, Ruolin
dc.contributor.author
Bukas, Christina
dc.contributor.author
de Andrade e Sousa, Lisa Barros
dc.contributor.author
Thalmeier, Dominik
dc.contributor.author
Baumann, Philipp F. M.
dc.contributor.author
Kook, Lucas
dc.contributor.author
Klein, Nadja
dc.contributor.author
Müller, Christian L.
dc.date.accessioned
2023-03-15T07:03:39Z
dc.date.available
2023-02-06T12:11:34Z
dc.date.available
2023-02-08T08:53:26Z
dc.date.available
2023-02-08T08:54:16Z
dc.date.available
2023-03-15T07:03:39Z
dc.date.issued
2023-01-18
dc.identifier.issn
1548-7660
dc.identifier.other
10.18637/jss.v105.i02
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/597165
dc.identifier.doi
10.3929/ethz-b-000597165
dc.description.abstract
In this paper we describe the implementation of semi-structured deep distributional regression, a flexible framework to learn conditional distributions based on the combination of additive regression models and deep networks. Our implementation encompasses (1) a modular neural network building system based on the deep learning library TensorFlow for the fusion of various statistical and deep learning approaches, (2) an orthogonalization cell to allow for an interpretable combination of different subnetworks, as well as (3) pre-processing steps necessary to set up such models. The software package allows to define models in a user-friendly manner via a formula interface that is inspired by classical statistical model frameworks such as mgcv. The package’s modular design and functionality provides a unique resource for both scalable estimation of complex statistical models and the combination of approaches from deep learning and statistics. This allows for state-of-the-art predictive performance while simultaneously retaining the indispensable interpretability of classical statistical models.
en_US
dc.format
application/pdf
en_US
dc.language.iso
en
en_US
dc.publisher
UCLA, Department of Statistics
en_US
dc.rights.uri
http://creativecommons.org/licenses/by/3.0/
dc.subject
Additive predictors
en_US
dc.subject
Deep learning
en_US
dc.subject
Effect decomposition
en_US
dc.subject
Orthogonal complement
en_US
dc.subject
Penalization
en_US
dc.subject
Smoothing
en_US
dc.title
deepregression: A Flexible Neural Network Framework for Semi-Structured Deep Distributional Regression
en_US
dc.type
Journal Article
dc.rights.license
Creative Commons Attribution 3.0 Unported
ethz.journal.title
Journal of Statistical Software
ethz.journal.volume
105
en_US
ethz.journal.issue
2
en_US
ethz.pages.start
1
en_US
ethz.pages.end
31
en_US
ethz.version.deposit
publishedVersion
en_US
ethz.identifier.wos
ethz.publication.place
Los Angeles, CA
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02120 - Dep. Management, Technologie und Ökon. / Dep. of Management, Technology, and Ec.::02525 - KOF Konjunkturforschungsstelle / KOF Swiss Economic Institute::06336 - KOF FB Data Science und Makroökon. Meth. / KOF FB Data Science and Macroec. Methods
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02120 - Dep. Management, Technologie und Ökon. / Dep. of Management, Technology, and Ec.::02525 - KOF Konjunkturforschungsstelle / KOF Swiss Economic Institute
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02120 - Dep. Management, Technologie und Ökon. / Dep. of Management, Technology, and Ec.::02525 - KOF Konjunkturforschungsstelle / KOF Swiss Economic Institute::06336 - KOF FB Data Science und Makroökon. Meth. / KOF FB Data Science and Macroec. Methods
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02120 - Dep. Management, Technologie und Ökon. / Dep. of Management, Technology, and Ec.::02525 - KOF Konjunkturforschungsstelle / KOF Swiss Economic Institute
ethz.tag
KOF-Key-refereed
en_US
ethz.date.deposited
2023-02-06T12:11:34Z
ethz.source
FORM
ethz.eth
yes
en_US
ethz.availability
Open access
en_US
ethz.rosetta.installDate
2023-02-08T08:53:28Z
ethz.rosetta.lastUpdated
2024-02-02T21:00:51Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=deepregression:%20A%20Flexible%20Neural%20Network%20Framework%20for%20Semi-Structured%20Deep%20Distributional%20Regression&rft.jtitle=Journal%20of%20Statistical%20Software&rft.date=2023-01-18&rft.volume=105&rft.issue=2&rft.spage=1&rft.epage=31&rft.issn=1548-7660&rft.au=R%C3%BCgamer,%20David&Kolb,%20Chris&Fritz,%20Cornelius&Pfisterer,%20Florian&Kopper,%20Philipp&rft.genre=article&rft_id=info:doi/10.18637/jss.v105.i02&
Files in this item
Publication type
-
Journal Article [133665]