Abstract
The Dutch government compensates health insurance companies when insuring
individuals who are estimated to have high health care costs. This is necessary
to avoid insurers not offering services to certain groups or not providing them
with a high quality of service. It is, however, unknown to what extent the
differences in health care expenses by different groups of people are truly due to
a poorer or better health status. We explore several statistical approaches that
facilitate explaining the cause of these differences.
individuals who are estimated to have high health care costs. This is necessary
to avoid insurers not offering services to certain groups or not providing them
with a high quality of service. It is, however, unknown to what extent the
differences in health care expenses by different groups of people are truly due to
a poorer or better health status. We explore several statistical approaches that
facilitate explaining the cause of these differences.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 126th European Study Group Mathematics with Industry |
| Subtitle of host publication | SWI 2017 |
| Editors | Daan Crommelin, Stella Kapodistria, Guus Regts, Chris Stolk, Peter van de Ven |
| Publisher | Studiegroep Wiskunde met de Industrie |
| Pages | 29-49 |
| Number of pages | 21 |
| Publication status | Published - Apr 2018 |
| Event | SWI 2017 - Amsterdam Duration: 23 Jan 2017 → 27 Jan 2017 |
Conference
| Conference | SWI 2017 |
|---|---|
| City | Amsterdam |
| Period | 23/01/17 → 27/01/17 |
Keywords
- health insurance
- risk equalisation
- model selection
- predictivemodel
- explanatory model
- lasso
- elastic net
- ridge regression
- clustering