An Approximation of the Non-Life Reserve Risk Distribution Using the Cornish-Fisher Expansion

23 Pages Posted: 10 May 2017

Date Written: May 3, 2014

Abstract

In order to get more insights into the characteristics of the non-life reserve risk distribution, the Cornish-Fisher expansion model is used. For this model, the knowledge of the first four moments of the distribution is necessary. These moments are derived using closed-form formulas. For the first two moments, the results described in Mack 1993 are used. For the third and fourth moments, closed-form models are introduced.

The proposed model is tested on 7 different triangles publicly available. It appears that the resulting distributions are different depending on the line of business but also on the (re)insurer. Therefore, it can be concluded that the knowledge of the characteristics of the reserve risk distribution can contribute to a better modelling of the reserve risk of each (re)insurer.

Keywords: Cornish-Fisher expansion, Kurtosis, Skewness, Variance, Chain-Ladder, Reserve risk distribution

JEL Classification: G22

Suggested Citation

Dal Moro, Eric, An Approximation of the Non-Life Reserve Risk Distribution Using the Cornish-Fisher Expansion (May 3, 2014). Available at SSRN: https://ssrn.com/abstract=2965384 or http://dx.doi.org/10.2139/ssrn.2965384

Eric Dal Moro (Contact Author)

Baloise ( email )

Basel, 4002
Switzerland

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
46
Abstract Views
473
PlumX Metrics