01.06.2015 Views

Actuarial Modelling of Claim Counts Risk Classification, Credibility ...

Actuarial Modelling of Claim Counts Risk Classification, Credibility ...

Actuarial Modelling of Claim Counts Risk Classification, Credibility ...

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

194 <strong>Actuarial</strong> <strong>Modelling</strong> <strong>of</strong> <strong>Claim</strong> <strong>Counts</strong><br />

Table 4.12 Linear relativities with risk classification for the system −1/ + 2<br />

and for Portfolio B.<br />

Level l Unconstrained relativities r l Linear relativities rl<br />

lin<br />

with a priori ratemaking<br />

with a priori ratemaking<br />

5 2232 % 2047%<br />

4 1713 % 1785%<br />

3 1489 % 1522%<br />

2 1121 % 1260%<br />

1 1050% 998%<br />

0 747% 736%<br />

The discrete approximations listed in Tables 3.5–3.6 can then be used in this formula.<br />

In the case where has a unimodal probability density function, the mixed uniform<br />

approximations <strong>of</strong> Tables 3.8–3.9 can also be used.<br />

4.6 Relativities with an Exponential Loss Function<br />

4.6.1 Bayesian Relativities<br />

This section proposes an asymmetric loss function with one parameter that reflects the<br />

severity <strong>of</strong> the bonus-malus system. In order to reduce the maluses obtained with a quadratic<br />

loss, keeping a financially balanced system, we resort on an exponential loss function. Such<br />

loss functions have been applied in Section 3.4 in the classical credibility setting. Our purpose<br />

here is to apply exponential loss functions to determine the optimal relativities.<br />

When using the exponential loss function, the goal is now to minimize<br />

[<br />

]<br />

Q exp = E exp−c − r L <br />

(4.18)<br />

under the financial balance constraint Er L = 1. The parameter c>0 determines the<br />

‘severity’ <strong>of</strong> the bonus-malus scale. The loss (4.18) puts more weight on the errors<br />

resulting in an overestimation <strong>of</strong> the premium (i.e. r L >), than on those coming from<br />

an underestimation. Consequently, the maluses are reduced, as well as the bonuses since<br />

financial stability has been imposed.<br />

Let us derive the general solution <strong>of</strong> (4.18).<br />

Proposition 4.1<br />

The solution <strong>of</strong> the constrained optimization problem (4.18) is<br />

r exp<br />

L<br />

= 1 + 1 c<br />

( [<br />

]<br />

)<br />

E ln Eexp−cL − ln Eexp−cL (4.19)<br />

Pro<strong>of</strong><br />

First, note that<br />

( [<br />

])<br />

exp ( ) expc exp E ln E exp−cL<br />

cr exp<br />

L =<br />

E [ exp−cL ]

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!