Final Exam: AK5182 Asuransi Umum (General Insurance)

15 Dec 2017

Q1. Losses have a lognormal distribution with 𝜇 = 7 and 𝜎 = 2. There is a deductible of 3,000 and 8
losses are expected each year. Determine the loss elimination ratio. If there is uniform inflation of 10%
but the deductible remains at 3,000 how many payment will be expected? Commented [HLL1]: 12/2008 Q5… a Pareto distribution
with 𝛼 = 3 and 𝜃 = 2000 deductible of 500 and 15 losses
Q2. In a particular year, let say in Year 1, total expected losses are 11,000,000. Individual losses in that
year have a Pareto distribution with 𝛼 = 3 and 𝜃 = 6,000. A reinsurer pays the excess of each individual
loss over 4,000. For this, the insurer is paid a premium equal to 115% of expected covered losses. One
year later, that is in Year 2, losses will experience 6% inflation over the previous year, but the frequency
of losses will not change. Determine the ratio of the premium in Year 2 to the premium in Year 1. Commented [HLL2]: 12/2008 Q1… losses are 10,000,000.
A lognormal distribution with 𝜇 = 8 and 𝜎 = 1 ...over 6,000,
Q3. Let an aggregate loss S follows a compound distribution. The primary distribution is Poisson with 3% inflation
mean 0.2 and the secondary distribution is 𝑈(0,5). If the policies are modified with a deductible of 𝑑 =
1 and a maximum covered loss of = 4, compute the mean and the variance of the modified aggregate
loss 𝑆.

Q4. An insurance portfolio produces N claims, where

N 𝑝𝑛 = 𝑃 [𝑁 = 𝑛]
0 0.5
1 0.4
3 0.1
Individual claim amounts have the following distribution:

X 𝑓𝑋 (𝑥)
1 0.9
10 0.1
Individual claim amounts and N are mutually independent.

(a) Determine the support (or sample space) of the aggregate claims 𝑆 = 𝑋1 + 𝑋2 + ⋯ + 𝑋𝑁
(b) Find the probability function of S
(c) Calculate the probability that the ratio of aggregate claims to expected claims 𝐸[𝑆] will exceed
3.0 Commented [HLL3]: 1/2010 Q5

Hint:

 𝑓𝑆 (𝑠) = ∑∞ ∗𝑛
𝑛=0 𝑝𝑛 𝑓𝑋 (𝑠) , 𝑛 = 0,1, …
∗(𝑛−1)
 𝑓𝑋∗𝑛 (𝑠) 𝑠
= ∑𝑦=0 𝑓𝑋 (𝑠 − 𝑦)𝑓𝑋 (𝑦) , 𝑛 = 2,3, …
 𝑓𝑋∗1 (𝑠) = 𝑓𝑋 (𝑠) ; 𝑓𝑋∗0 (0) = 1 ; 𝑓𝑋∗0 (𝑠) = 0, 𝑓𝑜𝑟 𝑠 ≠ 0

15 Dec 2016

Q1. Assume the aggregate loss 𝑆 follow a compound Poisson distribution with parameter 𝜆; with the
secondary distribution follows a Pareto distribution with parameter 𝛼 and 𝜃, where 𝛼 > 2.

(a) Determine the mean and the variance of 𝑆

Q2. Losses have a mean of 2000. With a deductible of 𝑑 = 1,000, the loss elimination ratio is 0.3. The
probability of a loss being greater than 1,000 is 0.4.

(a) Determine the average of a loss given it is less than or equal to 1,000
(b) Compute 𝐸[(𝑋 − 1,000)+ and the Mean Excess Loss 𝑒𝑋 (𝑑) = 𝐸[𝑋 − 𝑑|𝑋 > 𝑑]

Q3. Aggregate claims for a health plan have a Pareto distribution with 𝛼 = 2 and 𝜃 = 500. The health
plan implements an incentive to physicians that will pay a bonus of 50% of “the amount by which”
aggregate claims are less than 500, otherwise no bonus is paid. It is anticipated that with the incentive
plan, the claim distribution will change to become Pareto with 𝛼 = 2 and 𝜃 = 𝑐. With the new
distribution, it turns out that expected claims plus the expected bonus is equal to expected claims prior
to the bonus system. Determine the value of c.

Q4. A portfolio has 1000 independent policies. Each policy has a probability of 0.9 making no claim, and
a probability of 0.1 making a claim. Claim amount follows a Pareto distribution with 𝛼 = 3 and 𝜃 = 200.
Determine an approximate chance that the aggregate claim is larger than 14,900.

Q5. An insurer pays 80% of the aggregate loss in excess of the deductible of amount 5 up to a maximum
payment of 25. Aggregate claim amounts 𝑆 are integers and the following expected losses are known:
𝐸[(𝑆 − 5)+ = 8.52, 𝐸[(𝑆 − 36)+ = 3.25, and 𝐸[(𝑆 − 37)+ = 2.98. Calculate the expected amount paid
by the insurer.

17 Dec 2008

Q2. Let the distribution of aggregate claims 𝑆 follows a Pareto distribution with 𝛼 = 3 and 𝜃 = 1000.
The relative loading k and the value 𝜆 are selected so that Pr[𝑆 ≤ (1 + 𝑘)𝐸[𝑆]] = Pr[𝑆 ≤ (𝐸[𝑆] +
𝜆√𝑉𝑎𝑟[𝑆]) = 0.95. Calculate 𝑘 and 𝜆.

Q3. Let the distribution of number of losses follows a negative binomial distribution with parameters
𝑟 = 2 and 𝛽 = 3. Assume that losses follow a gamma distribution with 𝛼 = 200 and 𝜃 = 3

a. Let 𝑁 𝐿 denotes the number of losses and let 𝑁 𝑃 denotes the number of payments after a
deductible of d is imposed. Show that 𝑃𝑁𝑃 (𝑧) = 𝑃𝑁𝐿 [1 + 𝜗(𝑧 − 1)], where 𝜗 is the probability
when a payment is made
b. Examine the impact of a deductible of 350 has on the distribution of number of payments.

Q4. An insurer estimates that individual losses to an insured follow a lognormal distribution with
parameters 𝜇 = 7 and 𝜎 = 2. The insurer pays 75% of individual losses in excess of 2,000 with a
maximum payment of 150,000. It reinsures that portion of any payments in excess of 75,000.

a. Develop a model for the aggregate loss to the insurer after the reinsurance payment, where the
number of payments made by the insurer is denoted by the random variable N
b. If N follows a Poisson distribution with 𝜆 = 1.2, determine the mean and standard deviation of
the aggregate loss
6 Jan 2010