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

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Bonus-Malus Scales 173<br />

4.3.2 Transition Matrix<br />

Further, P is the one-step transition matrix, i.e.<br />

⎛<br />

p 00 p 01 ··· p 0s <br />

p 10 p 11 ··· p 1s <br />

P = ⎜<br />

⎝<br />

<br />

<br />

p s0 p s1 ··· p ss <br />

⎞<br />

⎟<br />

⎠ <br />

From (4.1), we see that the matrix P is a stochastic matrix. As already mentioned, the<br />

future level <strong>of</strong> a policyholder is independent <strong>of</strong> its past levels and only depends on its present<br />

level (and also on the number <strong>of</strong> claims reported during the present year).<br />

In matrix form, we can write P as<br />

P =<br />

∑<br />

k=0<br />

k<br />

exp −Tk<br />

k!<br />

provided the N t s are independent and oi distributed.<br />

Example 4.5 (−1/Top Scale)<br />

system is given by<br />

⎛<br />

P =<br />

⎜<br />

⎝<br />

The transition matrix P associated with this bonus-malus<br />

exp− 0 0 0 0 1− exp−<br />

exp− 0 0 0 0 1− exp−<br />

0 exp− 0 0 0 1− exp−<br />

0 0 exp− 0 0 1− exp−<br />

0 0 0 exp− 0 1− exp−<br />

0 0 0 0 exp− 1 − exp−<br />

Example 4.6 (−1/+2 Scale)<br />

system is given by<br />

⎞<br />

<br />

⎟<br />

⎠<br />

The transition matrix P associated with this bonus-malus<br />

⎛<br />

<br />

exp− 0 exp− 0<br />

2<br />

exp− 1 − ⎞<br />

2 1<br />

⎜ exp− 0 0 exp− 0 1− 2<br />

⎜⎜⎜⎜⎝ 0 exp− 0 0 exp− 1 − P =<br />

3<br />

0 0 exp− 0 0 1− exp−<br />

<br />

⎟<br />

0 0 0 exp− 0 1− exp− ⎠<br />

0 0 0 0 exp− 1 − exp−<br />

where i represents the sum <strong>of</strong> the elements in columns 1 to 5 in row i, i = 1 2 3, that is,<br />

)<br />

1 = exp−<br />

(1 + + 2<br />

2<br />

2 = 3 = exp− 1 +

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

Saved successfully!

Ooh no, something went wrong!