Insights from Inside Neural Networks
dc.contributor.author
Ferrario, Andrea
dc.contributor.author
Noll, Alexander
dc.contributor.author
Wüthrich, Mario V.
dc.date.accessioned
2020-11-05T13:25:11Z
dc.date.available
2019-01-16T09:01:54Z
dc.date.available
2019-03-11T14:31:49Z
dc.date.available
2020-11-03T10:11:39Z
dc.date.available
2020-11-05T13:25:11Z
dc.date.issued
2018-08-19
dc.identifier.other
10.2139/ssrn.3226852
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/316208
dc.description.abstract
We provide a tutorial that illuminates the aspects which need to be considered when fitting neural network regression models to claims frequency data in insurance. We discuss feature pre-processing, choice of loss function, choice of neural network architecture, class imbalance problem, as well as over-fitting. This discussion is based on a publicly available real car insurance data set.
en_US
dc.language.iso
en
en_US
dc.publisher
Social Science Research Network
en_US
dc.subject
Neural Networks
en_US
dc.subject
Architecture
en_US
dc.subject
Over-Fitting
en_US
dc.subject
Loss Function
en_US
dc.subject
Dropout
en_US
dc.subject
Regularization
en_US
dc.subject
LASSO
en_US
dc.subject
Ridge
en_US
dc.subject
Gradient Descent
en_US
dc.subject
Class Imbalance
en_US
dc.subject
Car Insurance
en_US
dc.subject
Claims Frequency
en_US
dc.subject
Poisson Regression Model
en_US
dc.subject
Machine Learning
en_US
dc.subject
Deep Learning
en_US
dc.title
Insights from Inside Neural Networks
en_US
dc.type
Working Paper
ethz.journal.title
SSRN
ethz.pages.start
3226852
en_US
ethz.size
52 p.
en_US
ethz.code.jel
JEL - JEL::G - Financial Economics::G2 - Financial Institutions and Services::G22 - Insurance; Insurance Companies; Actuarial Studies
en_US
ethz.code.jel
JEL - JEL::C - Mathematical and Quantitative Methods::C1 - Econometric and Statistical Methods and Methodology: General::C10 - General
en_US
ethz.code.jel
JEL - JEL::C - Mathematical and Quantitative Methods::C1 - Econometric and Statistical Methods and Methodology: General::C13 - Estimation: General
en_US
ethz.code.jel
JEL - JEL::C - Mathematical and Quantitative Methods::C1 - Econometric and Statistical Methods and Methodology: General::C14 - Semiparametric and Nonparametric Methods: General
en_US
ethz.code.jel
JEL - JEL::C - Mathematical and Quantitative Methods::C6 - Mathematical Methods; Programming Models; Mathematical and Simulation Modeling::C67 - Input–Output Models
en_US
ethz.publication.place
Rochester, NY
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02000 - Dep. Mathematik / Dep. of Mathematics::02003 - Mathematik Selbständige Professuren::08813 - Wüthrich, Mario Valentin (Tit.-Prof.)
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02000 - Dep. Mathematik / Dep. of Mathematics::02003 - Mathematik Selbständige Professuren::08813 - Wüthrich, Mario Valentin (Tit.-Prof.)
ethz.date.deposited
2019-01-16T09:02:05Z
ethz.source
FORM
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2019-03-11T14:31:59Z
ethz.rosetta.lastUpdated
2024-02-02T12:26:57Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Insights%20from%20Inside%20Neural%20Networks&rft.jtitle=SSRN&rft.date=2018-08-19&rft.spage=3226852&rft.au=Ferrario,%20Andrea&Noll,%20Alexander&W%C3%BCthrich,%20Mario%20V.&rft.genre=preprint&rft_id=info:doi/10.2139/ssrn.3226852&
Files in this item
| Files | Size | Format | Open in viewer |
|---|---|---|---|
|
There are no files associated with this item. |
|||
Publication type
-
Working Paper [5717]