1
Week 1 1
Principles of Actuarial Science;
Games of Chance (1)
Outline:
1. Traditional Actuarial Application: Insurance
2. Principles of Actuarial Science
3. Probability Basics (Revision?)
4. Poisson and Binomial Distributions
Reading
(Req) Sherris, 1.3-1.4, 2.1-2.3
(Recc) Sherris 1.1-1.2
1
Version: 15 July 2010. Copyright UNSW Actuarial Studies
2
Traditional Application of Actuarial Science - Insurance
What is insurance?
Examples:
– Life (Death)
– Life (Survival)
– Health
– Disability
– Motor
– House
– Home Contents
– Workers Compensation
Fundamental Idea: Payment(s) that are incurred when a (random) event occurs
Other actuarial applications: Superannuation, Risk Management
(Will discuss in much more detail later in the course)
3
Discussion
Consider a life insurance policy that:
– pays $100 at the end of the year if the policyholder ("Harry P") dies during the year
– assume that Harry has a 50% chance of dying
Q: Suppose that you are the actuary of the insurance company. Suggest how you
would determine the premium of this policy (making any additional assumptions if
required)
4
Basic Principles of Actuarial Science
Probability and Statistics
– quantify future uncertainty
– probability models - Binomial, Poisson, Normal, Log-Normal
– Survival models
Risk Pooling
– Independent Risks
– Risk Factors (age, sex, smoker status etc)
Law of Large Numbers
– Expected values
– Variability of averages less for pools of independent risks
5
Interest Rates
– Time Value of money
– economics of interest rates
– present values
Risk
– Utility Theory and Ranking of risky outcomes
– expected utility principle, certainty equivalents
6
Discussion (ctd)
Suppose the premium for the policy has been set at $55, and that interest rates are
0.
What are some of the practical issues the company must now consider? Consider:
–(i) What to do with the $55 received as a premium?
(ii) How should this policy be reflected in the company’s financial reports?
7
Basic Principles of Actuarial Science (ctd)
Financial Principles
– Accounting and Market values
– Fair Value of Liabilities
Actuarial Modelling
– Contingent Cash flows
– Death, survival, illness, financial asset returns
8
Risk Management
– What is risk?
– Pooling and Diversification of Risk
– Experience Rating
– Profit Sharing
– Solvency
Risk Classification
– Moral Hazard
Careless or deliberate actions of insured affect payments
– Adverse Selection
Worse than average risks have incentives to purchase insurance
– Information Asymmetry
9
Probability and Games of Chance
Probability
Permutation and Combinations
Useful Result: Binomial Theorem:
(x + y)n
=
n
X
r=0
n
r
xn r
yr
Simple method of determining probabilities (where/if possible!):
– Evaluate all possible equally likely cases
– Count the Frequency or proportion of times that the event occurs
10
Example
What is the probability of rolling a total of 12 with three dice?
11
Sum Frequency
3 1
4 3
5 6
6 10
7 15
8 21
9 25
10 27
11 27
12 25
13 21
14 15
15 10
16 6
17 3
18 1
Total 216
12
Random Variables
A random variable X is a function that assigns values to possible outcomes.
Discrete random variables - fixed number of finite values or countably infinite possible
values.
Continuous random variables - values on a continuous scale
Examples in Actuarial Science:
– Number of motor vehicle accidents occurring during a particular month
– Number of lives who die from cancer who are aged 40 last birthday during a par-
ticular time period.
– Time until the first accident for a particular car insurance policy
– Claim payment for a fire insurance policy
13
Probability Density and Distribution Functions
Probability a discrete random variable X takes a particular value x is denoted
f(x) = P(X = x)
Probability density has the properties
f(x) 0 and
X
all x
f(x) = 1
Distribution function, (= "cumulative distribution function") of X; is the probability that
the random variable takes a value less than or equal to a specified value x
F(x) = Pr (X x)
14
For continuous random variables the probability density function (p.d.f.) f(x) is de-
fined such that
Pr (a X b) =
Z b
a
f(x)dx
with
f(x) 0 for 1 < x < 1
Z 1
1
f(t)dt = 1
CDF:
F(x) = Pr (X x)
=
Z x
1
f(t)dt
15
Expectation
Discrete random variable - expectation of a function of a random variable g(X) is
defined as
E [g (X)] =
X
all x
g (x) f(x)
Continuous random variable
E [g (X)] =
Z 1
1
g (x) f(x)dx
16
Moments
Moments (about the origin) of a random variable are defined as
E [Xr
] r = 1; 2; : : :
First moment (r = 1) is the mean of the distribution of the random variable - usually
denoted by .
Discrete random variable
E [X] = =
X
all x
xf(x)
Continuous random variable
E [X] = =
Z 1
1
xf(x)dx
17
Moments - Variance
The mean is a measure of central tendency for a distribution.
The second moment about the mean is referred to as the variance of the distribution.
Variance usually denoted by 2
V ar [X] = 2
= E
h
(X )2
i
The standard deviation is the square root of the variance, .
The variance is a measure of dispersion or spread of the distribution.
18
Example
Consider a random variable X which takes the value 1 if Harry P dies and 0 if he
survives the year.. Find the mean, variance and standard deviation of X:
19
Common Distributions in Actuarial Science
Poisson
Binomial
Geometric
Exponential
Normal
Log-Normal
Weibull
20
Poisson Distribution
The Poisson distribution
- often used to model the probability of occurrence of rare events such as insurance
claims
- also as an approximation for the chance of dying for a small enough parameter .
Poisson ( )
Pr (X = x) =
e x
x!
x = 0; 1; 2 : : :
E [X] =
V ar [X] =
21
Exercise
Assume that the probability of an insurance claim on a particular insurance policy
during a year has a Poisson distribution with = 1
10:
Calculate the probability that there will be
(a) no claims during the year on the policy?
(b) exactly 2 claims?
(c) at least one claim?
22
Binomial Distribution
Number of successes (x) in a series of n independent trials where each success has
probability p.
In actuarial science
– probability that a number of lives in a group of lives with the same chance of dying
will die assuming independence between the lives.
Binomial (n; p)
Pr (X = x) =
n
x
px
(1 p)n x
x = 0; 1; 2 : : :
E [X] = np
V ar [X] = np (1 p)
23
Example
Suppose we have 100 policyholders, each with a probability of 0.2 of dying within the
next year. What is the probability that we observe a total of 10 or less deaths by the
end of the year?

Principles of Actuarial Science Chapter 1

  • 1.
    1 Week 1 1 Principlesof Actuarial Science; Games of Chance (1) Outline: 1. Traditional Actuarial Application: Insurance 2. Principles of Actuarial Science 3. Probability Basics (Revision?) 4. Poisson and Binomial Distributions Reading (Req) Sherris, 1.3-1.4, 2.1-2.3 (Recc) Sherris 1.1-1.2 1 Version: 15 July 2010. Copyright UNSW Actuarial Studies
  • 2.
    2 Traditional Application ofActuarial Science - Insurance What is insurance? Examples: – Life (Death) – Life (Survival) – Health – Disability – Motor – House – Home Contents – Workers Compensation Fundamental Idea: Payment(s) that are incurred when a (random) event occurs Other actuarial applications: Superannuation, Risk Management (Will discuss in much more detail later in the course)
  • 3.
    3 Discussion Consider a lifeinsurance policy that: – pays $100 at the end of the year if the policyholder ("Harry P") dies during the year – assume that Harry has a 50% chance of dying Q: Suppose that you are the actuary of the insurance company. Suggest how you would determine the premium of this policy (making any additional assumptions if required)
  • 4.
    4 Basic Principles ofActuarial Science Probability and Statistics – quantify future uncertainty – probability models - Binomial, Poisson, Normal, Log-Normal – Survival models Risk Pooling – Independent Risks – Risk Factors (age, sex, smoker status etc) Law of Large Numbers – Expected values – Variability of averages less for pools of independent risks
  • 5.
    5 Interest Rates – TimeValue of money – economics of interest rates – present values Risk – Utility Theory and Ranking of risky outcomes – expected utility principle, certainty equivalents
  • 6.
    6 Discussion (ctd) Suppose thepremium for the policy has been set at $55, and that interest rates are 0. What are some of the practical issues the company must now consider? Consider: –(i) What to do with the $55 received as a premium? (ii) How should this policy be reflected in the company’s financial reports?
  • 7.
    7 Basic Principles ofActuarial Science (ctd) Financial Principles – Accounting and Market values – Fair Value of Liabilities Actuarial Modelling – Contingent Cash flows – Death, survival, illness, financial asset returns
  • 8.
    8 Risk Management – Whatis risk? – Pooling and Diversification of Risk – Experience Rating – Profit Sharing – Solvency Risk Classification – Moral Hazard Careless or deliberate actions of insured affect payments – Adverse Selection Worse than average risks have incentives to purchase insurance – Information Asymmetry
  • 9.
    9 Probability and Gamesof Chance Probability Permutation and Combinations Useful Result: Binomial Theorem: (x + y)n = n X r=0 n r xn r yr Simple method of determining probabilities (where/if possible!): – Evaluate all possible equally likely cases – Count the Frequency or proportion of times that the event occurs
  • 10.
    10 Example What is theprobability of rolling a total of 12 with three dice?
  • 11.
    11 Sum Frequency 3 1 43 5 6 6 10 7 15 8 21 9 25 10 27 11 27 12 25 13 21 14 15 15 10 16 6 17 3 18 1 Total 216
  • 12.
    12 Random Variables A randomvariable X is a function that assigns values to possible outcomes. Discrete random variables - fixed number of finite values or countably infinite possible values. Continuous random variables - values on a continuous scale Examples in Actuarial Science: – Number of motor vehicle accidents occurring during a particular month – Number of lives who die from cancer who are aged 40 last birthday during a par- ticular time period. – Time until the first accident for a particular car insurance policy – Claim payment for a fire insurance policy
  • 13.
    13 Probability Density andDistribution Functions Probability a discrete random variable X takes a particular value x is denoted f(x) = P(X = x) Probability density has the properties f(x) 0 and X all x f(x) = 1 Distribution function, (= "cumulative distribution function") of X; is the probability that the random variable takes a value less than or equal to a specified value x F(x) = Pr (X x)
  • 14.
    14 For continuous randomvariables the probability density function (p.d.f.) f(x) is de- fined such that Pr (a X b) = Z b a f(x)dx with f(x) 0 for 1 < x < 1 Z 1 1 f(t)dt = 1 CDF: F(x) = Pr (X x) = Z x 1 f(t)dt
  • 15.
    15 Expectation Discrete random variable- expectation of a function of a random variable g(X) is defined as E [g (X)] = X all x g (x) f(x) Continuous random variable E [g (X)] = Z 1 1 g (x) f(x)dx
  • 16.
    16 Moments Moments (about theorigin) of a random variable are defined as E [Xr ] r = 1; 2; : : : First moment (r = 1) is the mean of the distribution of the random variable - usually denoted by . Discrete random variable E [X] = = X all x xf(x) Continuous random variable E [X] = = Z 1 1 xf(x)dx
  • 17.
    17 Moments - Variance Themean is a measure of central tendency for a distribution. The second moment about the mean is referred to as the variance of the distribution. Variance usually denoted by 2 V ar [X] = 2 = E h (X )2 i The standard deviation is the square root of the variance, . The variance is a measure of dispersion or spread of the distribution.
  • 18.
    18 Example Consider a randomvariable X which takes the value 1 if Harry P dies and 0 if he survives the year.. Find the mean, variance and standard deviation of X:
  • 19.
    19 Common Distributions inActuarial Science Poisson Binomial Geometric Exponential Normal Log-Normal Weibull
  • 20.
    20 Poisson Distribution The Poissondistribution - often used to model the probability of occurrence of rare events such as insurance claims - also as an approximation for the chance of dying for a small enough parameter . Poisson ( ) Pr (X = x) = e x x! x = 0; 1; 2 : : : E [X] = V ar [X] =
  • 21.
    21 Exercise Assume that theprobability of an insurance claim on a particular insurance policy during a year has a Poisson distribution with = 1 10: Calculate the probability that there will be (a) no claims during the year on the policy? (b) exactly 2 claims? (c) at least one claim?
  • 22.
    22 Binomial Distribution Number ofsuccesses (x) in a series of n independent trials where each success has probability p. In actuarial science – probability that a number of lives in a group of lives with the same chance of dying will die assuming independence between the lives. Binomial (n; p) Pr (X = x) = n x px (1 p)n x x = 0; 1; 2 : : : E [X] = np V ar [X] = np (1 p)
  • 23.
    23 Example Suppose we have100 policyholders, each with a probability of 0.2 of dying within the next year. What is the probability that we observe a total of 10 or less deaths by the end of the year?