0% found this document useful (0 votes)
15 views15 pages

Aggregate Risk Models in Insurance

The presentation discusses aggregate risk models in insurance, focusing on basic terminologies such as loss, claim, frequency, severity, and aggregate loss. It explains individual and collective risk models, providing examples and key properties, including the use of compound distributions. Additionally, it presents a case study on liability motor insurance, estimating the mean and standard deviation of aggregate loss using simulated data.

Uploaded by

soostutimilsina
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views15 pages

Aggregate Risk Models in Insurance

The presentation discusses aggregate risk models in insurance, focusing on basic terminologies such as loss, claim, frequency, severity, and aggregate loss. It explains individual and collective risk models, providing examples and key properties, including the use of compound distributions. Additionally, it presents a case study on liability motor insurance, estimating the mean and standard deviation of aggregate loss using simulated data.

Uploaded by

soostutimilsina
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd

Aggregate Risk Models

A Presentation in Actuarial Seminar

Swastika Timilsina
Roll No: 28

School of Mathematical Sciences


Balkhu
2025/8/28 1
Basic Terminologies

Loss: The money lost by the insured

Claim: Payment made when an insured event happens

Frequency: How often an insured event occurs (claim count) in a period


(like 6 months or a year).

Severity: The size or amount of a single loss

Aggregate Loss: The total amount paid for all insureds in a period.
2
Aggregate Risk
is the random variable representing the total amount of claims payable
by a company in a relatively short fixed period of time from a portfolio
or collection of policies.

Restricting consideration to shorter periods of time like a few months


or a year often allows to ignore aspects of the changing value of money
due to inflation.

3
Types of Risk Model
Individual Risk Model: record losses for each contract and then
add them up.

Collective Risk Model (compound model): record losses


as claims are made and then add them up.

4
Example:
An insurance portfolio of four policies:
Policy ID Claim ID Loss Amount
1 - -
2 1 10
3 1 10
3 2 10
4 1 10
4 2 10
4 3 10

Aggregate Losses:
Individual risk model: 0 + 10 + 20 + 30 = 60
Collective risk model: 10 + 10 + 10 + 10+ 10+ 10 = 60
5
Individual Risk Models
The individual risk model represents the aggregate loss as a sum
of a fixed number of insurance contracts

where
denotes the aggregate loss for n (a fixed number) contracts
denotes the loss for the contract for
are assumed to independent but are not necessarily identically
distributed, due to different coverage or exposure
usually has a probability mass at zero
6
Where
indicated by 1 and 0 with probability . It indicates whether the policy
has a claim.

represents the amount of losses of policy i, given that a claim is made.

7
Basic Properties

Where
= probability of claim from policy
= Amount of claim from policy

8
Collective Risk Model
The collective risk model has representation

with S being the aggregate loss of N (a random number) individual


claims
Key assumptions
• are i.i.d. random variables
• The distribution of N and the common distribution of are
independent of each other.
Two building blocks: frequency and severity
9
Basic Properties

The distribution of is called the Compound Distribution.


Common Distributions of S are:
Compound Poisson Distribution
Compound Binomial Distribution
Compound Negative Binomial Distribution

10
Compound approximation for
individual risk models
In approximation:
1. Instead of modelling each person separately, we model claim count
and claim severity.
2. The claim count distribution is chosen so that (same expected
number of claims).
3. The severity distribution is chosen so that its mean matches the
weighted average of individual claim means.

11
Why are they not exactly equal?
• In the individual mode, each person can have at most one claim in the
period.
• In the collective model, the claim count N can assign multiple claims
to the same person.
• This difference means:
• The means match exactly
• The variances differ slightly — the compound model’s variance is a
bit larger because it allows extra variability from possible multiple
claims per individual.

12
Consider an insurance company that sells liability motor insurance
where an individual’s claim frequency, N, follows a Poisson
distribution with mean λ = 25 and their claim severity, X, follows the
Gamma distribution with shape parameter α = 5 and scale parameter
β = 300. Find mean and standard deviation of aggregate loss.
Solution:

13
Using a simulated sample of 10,000 observations, we could estimate
the mean and variance of the aggregate loss S.

The simulated mean aggregate loss is $37,374.97 with a standard


deviation of $8,186.02. 14
Thank You!

15

You might also like