umu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Assessing the merits of penalized regression in insurance pricing
Umeå University, Faculty of Science and Technology, Department of Mathematics and Mathematical Statistics.
2017 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Generalized linear models (GLMs) are global market standard for technical pricing models in insurance but have limited capabilities when modeling granular data such as vehicles or geography. Penalized regression models are extensions of GLMs and an emerging area of research in insurance pricing that could potentially compensate for these weaknesses and generate better results than existing approaches that involve complex analysis outside of its framework. This thesis aims to develop a penalized regression approach using granular data. We construct a fair comparison between a penalized regression model with granular car model data and the GLM in Willis Towers Watson's P&C pricing software Emblem. By visualizing double lift charts, we found that the penalized model is closer to the true risk than the GLM. The sum of squares is also 70% lower compared to the GLM. These results indicate that a penalized regression model can capture meaningful information in the granular car model data, and could probably in other granular risk factors as well.

Place, publisher, year, edition, pages
2017.
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:umu:diva-143990OAI: oai:DiVA.org:umu-143990DiVA, id: diva2:1174852
Examiners
Available from: 2018-02-05 Created: 2018-01-16 Last updated: 2018-02-05Bibliographically approved

Open Access in DiVA

No full text in DiVA

By organisation
Department of Mathematics and Mathematical Statistics
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 120 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf