Why using a general model in Solvency II is not a good idea : an explanation from a Bayesian point of view
The passing of Directive 2009/138/CE (Solvency II) has opened a new era in the European insurance market. According to this new regulatory environment, the volume of own resources will be determined depending on the risks that any insurer would be holding. So, nowadays, the model to estimate the amount of economic capital is one of the most important elements. The Directive establishes that the European entities can use a general model to perform these tasks. However, this situation is far from being optimal because the calibration of the general model has been made using figures that reflects and average behaviour. This paper shows that not all the companies operating in a specific market has the same risk profile. For this reason, it is unsatisfactory to use a general model for all of them. We use the PAM clustering method and afterwards some Bayesian tools to check the results previously obtained. Analysed data (public information belonging to Spanish insurance companies about balance sheets and income statements from 1998 to 2007) comes from the DGSFP (Spanish insurance regulator).
|Date of creation:||Nov 2011|
|Contact details of provider:|| Web page: http://portal.uc3m.es/portal/page/portal/dpto_estadistica|
References listed on IDEAS
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Hadfield, Jarrod D., 2010. "MCMC Methods for Multi-Response Generalized Linear Mixed Models: The MCMCglmm R Package," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 33(i02).
- J. David Cummins & Mary A. Weiss, 1991. "The structure, conduct, and regulation of the property-liability insurance industry," Conference Series ; [Proceedings], Federal Reserve Bank of Boston, vol. 35, pages 117-164.
- McCarty, John A. & Hastak, Manoj, 2007. "Segmentation approaches in data-mining: A comparison of RFM, CHAID, and logistic regression," Journal of Business Research, Elsevier, vol. 60(6), pages 656-662, June.
- Van Gestel, Tony & Martens, David & Baesens, Bart & Feremans, Daniel & Huysmans, Johan & Vanthienen, Jan, 2007. "Forecasting and analyzing insurance companies' ratings," International Journal of Forecasting, Elsevier, vol. 23(3), pages 513-529.
- A. N. Pettitt & T. T. Tran & M. A. Haynes & J. L. Hay, 2006. "A Bayesian hierarchical model for categorical longitudinal data from a social survey of immigrants," Journal of the Royal Statistical Society Series A, Royal Statistical Society, vol. 169(1), pages 97-114.
- Constantin Anghelache & Dan Armeanu, 2008. "Application of Discriminant Analysis on Romanian Insurance Market," Theoretical and Applied Economics, Asociatia Generala a Economistilor din Romania - AGER, vol. 11(11(528)), pages 51-62, November.
- Christophe Genolini & Bruno Falissard, 2010. "KmL: k-means for longitudinal data," Computational Statistics, Springer, vol. 25(2), pages 317-328, June.
- de Haan, Leo & Kakes, Jan, 2010.
"Are non-risk based capital requirements for insurance companies binding?,"
Journal of Banking & Finance,
Elsevier, vol. 34(7), pages 1618-1627, July.
- Leo de Haan & Jan Kakes, 2007. "Are non-risk based capital requirements for insurance companies binding?," DNB Working Papers 145, Netherlands Central Bank, Research Department.
- Dedu, Vasile & Armeanu, Daniel & Enciu, Adrian, 2009. "Using the Multivariate Data Analysis Techniques on the Insurance Market," Journal for Economic Forecasting, Institute for Economic Forecasting, vol. 0(4), pages 170-179, December.
When requesting a correction, please mention this item's handle: RePEc:cte:wsrepe:ws113729. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Ana Poveda)
If references are entirely missing, you can add them using this form.