LOSS MODEL 1
Maria Irsan
LEARNING CONTRACT
• Scoring Component
• Syllabus
• References
• Class Captain
SCORING COMPONENT
❑ Attendance 10%
❑ Quiz 15%
❑ Individual & Group Assignment 20%
❑ MID Examination 25%
❑ Final Examination
30%
TOPIC
• Modeling and Review Random Variable
• Basic Distributional Quantities (Moment, Percentiles, Generating Function and
Sum of Variables
• Tail of Distribution and Measure of Risk
• Characteristic of Actuarial Model
• Creating New Distribution
• Selected Distribution and Their Relationships
• Discrete Distribution: (a, b, 0) and (a, b, 1) class
• Compound Frequency Distribution
• Mixed Frequency Distribution
REFERENCE
Klugman, S.A., Panjer, H.H. and Willmot, G.E. (2019)
Loss Models: From Data to Decisions. Fifth Edition,
Wiley, New York.
General Insurance
MARIA YUS TRINITY IRSAN
Outline
u The model-based approach
u Course Description
u Random Variables - Introduction
u Random Variables – Key Function and 4 Models
Modeling
u The model-based Approach
math actuarial parame- Model
model problem ters selection
Modeling
u The Modeling Process
• Stage 1: one or
models are
selected
• Stage 2: The
model is
calibrated
• Stage 3:The
fitted model is
validated
Modeling
u The Modeling Process
• Stage 4:
Consider other
possible models
• Select the best
model
• Apply the
selected model
Modeling
u The Modeling Advantage
Decision making based strictly upon empirical
evidence.
*)Empirical approach assumes that the future can
be expected to be exactly like a sample from the
past.
Modeling
u Example
A portfolio of group life insurance certificates
consists of 1,000 employees of various ages and
death benefits. Over the past five years, 14
employees died and received a total of 580,000 in
benefits (adjusted for inflation because the plan
relates benefits to salary). Determine the empirical
estimate of next year's expected benefit payment
Course Description
u Topic 1: Review of probability
u Topic 2: Understanding probability distribution
u Topic 3: Coverage modification
u Topic 4: Aggregate Losses (optional)
Random Variables (Review)
Pay-
When ment
? ?
Actuarial seve-
Model rity?
Uncertain Stream of
Future Payment
Random Variables (Review)
u Definition of random variable and important functions
with some examples.
u Basic calculations from probability models.
u Specific probability distributions and their properties.
u More advanced calculations using severity models.
u Models incorporating the possibility of a random number
of payments each of random amount
Random Variables (Review)
Event
Possible
Outcome Random
Phenomena
Probability
of the
outcomes/
event
Random Variables (Review)
Phenomena are occurrences that can be observed.
An experiment is an observation of a given phenomenon
under specified conditions. The result of an experiment is called
an outcome; an event is a set of one or more possible
outcomes. A stochastic phenomenon is a phenomenon for
which an associated experiment has more than one possible
outcome. An event associated with a stochastic phenomenon
is said to be contingent. ... Probability is a measure of the
likelihood of the occurrence of an event, measured on a scale
of increasing likelihood from zero to one. ... A random variable
is a function that assigns a numerical value to every possible
outcome.
Random Variables (Review)
Example 12 random variables in ACS
u The age at death of a randomly selected birth.
u The time to death from when insurance was purchased for a
randomly selected insured life.
u The time from occurrence of a disabling event to recovery or death
for a randomly selected workers compensation claimant.
u The time from the incidence of a randomly selected claim to its
being reported to the insurer
u The time from the reporting of a randomly selected claim to its
settlement
Random Variables (Review)
Example 12 random variables in ACS
u The number of dollars paid on a randomly selected life insurance
claim
u The number of dollars paid on a randomly selected automobile
bodily injury claim
u The number of automobile bodily injury claims in one year from a
randomly selected insured automobile
u The total dollars in medical malpractice claims paid in one year
owing to events at a randomly selected hospital
Random Variables (Review)
Example 12 random variables in ACS
u The time to default or prepayment on a randomly selected insured
home loan that terminates early
u The amount of money paid at maturity on a randomly selected
high-yield bond
u The value of a stock index on a specified future date
Key Functions and Four Models
u Five Key Functions used in describing a random variable.
CDF Survival PDF
Prob. Hazard
Function Rate
Key Functions and Four Models
u The cumulative distribution function, also called the distribution
function and usually denoted !" # or ! # for a random variable X
is the probability that X is less than or equal to a given number. That
is, F% # = '( ) ≤ # .
u The cdf must satisfy:
1. 0 ≤ ! # ≤ 1, ∀#
2. ! # is nondecreasing
3. ! # is right-continuous
4. lim6→89 ! # = 0 and lim6→9 ! # = 1
Key Functions and Four Models
u Model 1
This random variable could serve as a
model for the age at death. All ages
between 0 and 100 are possible.
While experience suggests that there
is an upper bound for human lifetime,
models with no upper limit may be
useful if they assign extremely low
probabilities to extreme ages. This
allows the modeler to avoid setting a
specific maximum age:
Key Functions and Four Models
u Model 2
This random variable could serve as a
model for the number of dollars paid
on an automobile insurance claim. All
positive values are possible. As with
mortality, there is likely an upper limit
(all the money in the world comes to
mind), but this model illustrates that, in
modelling, correspondence to reality
need not be perfect:
Key Functions and Four Models
u Model 3
This random variable could serve as a model for the number of claims on one
policy in one year. Probability is concentrated at the five points (0,1,2,3,4) and
the probability at each is given by the size of the jump in the distribution
function:
Key Functions and Four Models
u Model 4
This random variable could serve as a model for the total dollars paid on a
medical malpractice policy in one year. Most of the probability is at zero (0.7)
because in most years nothing is paid. The remaining 0.3 of probability is
distributed over positive values:
Key Functions and Four Models
u The support of a random variable is the set of
numbers that are possible values of the random
variable.
u A random variable is called discrete if the support
contains at most a countable number of values. It
is called continuous if the distribution function is
continuous and is differentiable everywhere with
the possible exception of a countable number of
values. It is called mixed if it is not discrete and is
continuous everywhere with the exception of at
least one value and at most a countable number
Key Functions and Four Models
For each of the four models, determine the
support and indicate which type of random
variable it is.
u Model 1: Support: values from 0 to 100. Type:
continuous (because the support is continuous
and its differentiable everywhere except at 0
and 100 (countable)).
u Model 2, 3, 4?
Key Functions and Four Models
u The survival function, usually denoted !" # or
! # , for a random variable X is the probability
that X is greater than a given number. That is
!" # = Pr ' > # = 1 − +" # .
u As result
1. 0 ≤ ! # ≤ 1, ∀#
2. S # is noninreasing
3. S # is right-continuous
4. lim ! # = 1 and lim ! # = 0
5→78 5→8
Key Functions and Four Models
For each of the four models, determine the survival
function
u Model 1: S" # = 1 − 0.01#, 0 ≤ # < 100
u Model 2, 3, 4?
Key Functions and Four Models
For each of the four models, determine the density
function
u Model 1: !" # = % & # = 0.01, 0 < # < 100
u Model 2, 3, 4?
Key Functions and Four Models
u The probability function, also called the
probability mass function and usually denoted
!" # or ! # , describes the probability at a
distinct point when it is not 0. The formal
definition is !" # = Pr ' = # .
u For discrete random variables, the distribution
and survival functions can be recovered as
( # = Σ*+, ! - and . # = Σ*/, ! -
Key Functions and Four Models
For each of the four models, determine the
probability function
Key Functions and Four Models
u The hazard rate, also known as the force of
mortality and the failure rate and usually denoted ℎ" #
or ℎ # , is the ratio of the density and survival functions
% '
when the density function is defined. That is, ℎ" # = (& '
.
&
u Force of mortality denoted by ) # , failure rate denoted
by * #
(, ' - 6
3 ∫5 7 ' -'
u ℎ" # = −(' = − -' ln 0 # , implies 0 1 = 2
Key Functions and Four Models
For each of the four models, determine the hazard
rate.
%.%"
u Model 1: ℎ" # = , 0 < # < 100
"'%.%"(
u ℎ- , ℎ. , ℎ/ ???
Key Functions and Four Models
u Model 5
An alternative to the simple lifetime distribution in Model 1
Key Functions and Four Models
u The mode of a random variable is the most likely value. For a
discrete variable it is the value with the largest probability. For a
continuous variable it is the value for which the density function is
largest. If there are local maxima, these points are also considered
to be modes.
u Where possible, determine the mode for Models 1-5.
u Model 1: the pdf is constant, so all values could be a node or could
be said there is no mode
u Model 2, 3, 4, 5?
Exercise
u Determine cdf, pdf, and hazard rate function for Model 5.
u Construct graphs of cdf, pdf or probability function, hazard rate
function for Model 2, 3, 4, 5. (if it exists)
u A nonnegative random variable has hazard rate function of ℎ " =
$ + & '( , " ≥ 0. Given , 0.4 = 0.5. Find the value of A.
