Part II: Aggregate Loss Models

Samuel Hao (University of Manitoba) ACT 4020 47 / 94

Chapter 9: Aggregate Loss Models

Definition 9.1
The collective risk model has the representation

S = X1 + X2 + · · · + XN ,

where Xj ’s are individual payment amounts, and N is the number of

payments. This is a compound sum. Standard independence
assumptions are:
1 the random variables X1 , X2 , . . ., are i.i.d.;
2 the random variables X1 , X2 , . . ., are independent of the random
variable N.

Samuel Hao (University of Manitoba) ACT 4020 48 / 94

9.3 The compound model for aggregate claims
For a collective risk model,

FS (x) = Pr(S  x)
•
= Â pn Pr(X1 + · · · + Xn  x|N = n)
n=0
•
= Â pn Pr(X1 + · · · + Xn  x)
n=0
•
= Â pn FX⇤n (x)
n=0

Samuel Hao (University of Manitoba) ACT 4020 49 / 94

“n-fold convolution”:

If X is a non-negative continuous r.v.,

Z x
⇤k ⇤(k 1)
FX (x) = FX (x y)fX (y)dy, k = 2, 3, . . .
0

If X is discrete on non-negative integers,

x
⇤(k 1)
⇤k
FX (x) = Â FX (x y ) fX ( y ) , x = 0, 1, . . . , k = 2, 3, . . .
y=0

Samuel Hao (University of Manitoba) ACT 4020 50 / 94

Example
Suppose Xi , i = 1, 2, 3, are i.i.d. random variables with the following
distribution: 8
< 0 with p0
Xi = 1 with p1
:
2 with 1 p0 p1
Calculate Pr(X1 + X2 + X3  4).

Samuel Hao (University of Manitoba) ACT 4020 51 / 94

 E(S) = E(X ) · E(N )

Var(S) = E(N ) · Var(X) + [E(X)]2 · Var(N )

Samuel Hao (University of Manitoba) ACT 4020 52 / 94

Examples 9.3-9.4
The observed mean (and standard deviation) of the number of claims
and the individual losses over the past 10 months are 6.7 (2.3) and
179,247 (52,141), respectively. Using normal distributions as
approximating distributions for aggregate claims, calculate the
probability that claims will exceed 140% of expected costs.

Samuel Hao (University of Manitoba) ACT 4020 53 / 94

 Definition 9.3
Insurance on the aggregate losses, subject to a deductible, is called
stop-loss insurance. The expected cost of this insurance is called the
net stop-loss premium and can be computed as E[(S d)+ ], where d is
the deductible.

Z •
E [(S d) + ] = [1 FS (x)] dx
Zd •
= (x d) fS (x)dx
FE )
d
if pra < s < s) =
0 ,

√
,

的 ; bg
,

E] Sa
0
(s a
E[(s } s ( t ) dt
[
Jas
a)
]
=

E [ s b]
-

dt
-

+
Sca 3 < b
-

+
sll =
. -
a)

= ( b a ) S (a ]
-
.
E [ Ls - fydestr orS ( b ) t 的 -
G)

= ( s -
a ) -

S ( b)

Samuel Hao (University of Manitoba) ACT 4020 54 / 94

「
爪一
Exercise 9.16
A reinsurer pays aggregate claim amounts in excess of d, and in return
it receives a stop-loss premium E[(S d)+ ]. You are given
E[(S 100)+ ] = 15,
E[(S 120)+ ] = 10,
the probability that the aggregate claim amounts are greater than
80 and less than or equal to 120 is 0.
Determine the probability that the aggregate claims amounts are less
than or equal to 80.

Samuel Hao (University of Manitoba) ACT 4020 55 / 94

Exercise 9.14
For a group health contract, aggregate claims are assumed to have
an exponential distribution where the mean of the distribution is
estimated by the group underwriter.
Aggregate stop-loss insurance for total claims in excess of 125% of
the expected claims is provided by a gross premium that is twice
the expected stop-loss premium.
You have discovered an error in the underwriter’s method of
calculating expected claims. The underwriter’s estimate is 90% of
the correct estimate.
Determine the actual percentage loading in the premium.

Samuel Hao (University of Manitoba) ACT 4020 56 / 94

9.8 The individual risk model

Definition 9.2
When the random variable N in the collective risk model reduces to a
nonrandom constant n, then the aggregate losses have a representation

S = X1 + X2 + · · · + Xn ,

and the model is then called the individual risk model. Note that X1 ,
X2 , . . . are assumed to be independent but are not assumed to be
identically distributed.

Example: losses of a group life or health insurance policy that covers a

group of n employees.

Samuel Hao (University of Manitoba) ACT 4020 57 / 94

In most cases, S = I1 B1 + · · · + In Bn .
Ij is an indicator variable.

Bj is the amount of payment from the jth policy given that a

payment is made.

I1 , . . . , In , B1 , . . . , Bn are independent.

Samuel Hao (University of Manitoba) ACT 4020 58 / 94

Use a parametric distribution (normal, lognormal, gamma, etc.) to
approximate the distribution of S.

Exercise 9.69
A group life insurance contract covering independent lives is rated in
the three age groupings as given in Table 9.19. The insurer prices the
contract so that the probability that claims will exceed the premium is
0.05. Using the normal approximation, determine the premium that
the insurer will charge.

Samuel Hao (University of Manitoba) ACT 4020 59 / 94