General Insurance 2

GENERAL INSURANCE
MEETING 1
TOPICS
1. REVIEW : RANDOM VARIABLES
2. BASIC DISTRIBUTIONAL QUANTITES
3. CLASSIFYING AND CREATING DISTRIBUTION
4. FREQUENCY AND SEVERITY WITH COVERAGE MODIFICATIONS
5. AGGREGATE LOSS MODELS
References :
1. Stuart Klugman. Loss Models : From Data to Decisions. 2nd Edition. 2008
2. Yiu-Kuen Tse. Nonlife Actuarial Models : Theory, Methods, and
Evaluation.
RANDOM VARIABLES
(Cumulative Distribution Function)
Definition 1 :
The cumulative distribution function, alse called distribution function,
and usually denoted 𝐹𝑋 (𝑥) or 𝐹(𝑥), for random variable 𝑋 is the
probability that 𝑋 is less than or equal to a given number. That is,
𝐹𝑋 𝑥 = Pr(𝑋 ≤ 𝑥) . The abbreviation cdf is often used.
The distribution function must satisfy a number of requirements :
•0≤𝐹 𝑥 ≤1
• 𝐹(𝑥) is nondecreasing
• 𝐹(𝑥) is right-continous
• lim 𝐹 𝑥 = 0 and lim 𝐹 𝑥 = 1
𝑥→−∞ 𝑥→ ∞
CUMULATIVE DISTRIBUTION FUNCTION
Model 1.
This random variable could serve as a model for the age at death.
Model 2.
This random variable could serve as a model for the number of dollar
paid on an automobile insurance claim.
Model 3.
This random variable could serve as a model for the number of claims
on one policy in one year.
Model 4.
This random variable could serve as a model for the total dollars paid
on a malpractice policy in one year.
THE SURVIVAL FUNCTION
Definition 2 :
The survival function, usually denoted 𝑆𝑋 (𝑥) or 𝑆(𝑥), for a random
variable 𝑋 is the probability that 𝑋 is greater than a given number. That
is, 𝑆𝑋 𝑥 = Pr 𝑋 > 𝑥 = 1 − 𝐹𝑋 (𝑥).
As a result :
•0≤𝑆 𝑥 ≤1
• 𝑆(𝑥) is nonincreasing
• 𝑆(𝑥) is right-continous
• lim 𝑆 𝑥 = 1 and lim 𝑆 𝑥 = 0
𝑥→−∞ 𝑥→ ∞
Here are the survival functions for the four models :
Figure 1. Survival Function for Model 1
Figure 2. Survival function for model 2.

THE PROBABILITY DENSITY FUNCTION
Definition 3:
The probability density function, also called the density function,
usually denoted 𝑓𝑋 𝑥 , or 𝑓(𝑥), is the derivative of the distribution
function or, equivalently, the negative of the derivative of the survival
function.
For example : For model 1 : 𝑓𝑋 𝑥 = 0.01 , 0 ≤ 𝑥 < 100.
THE PROBABILITY DENSITY FUNCTION
The density function for our models :
PROBABILITY FUNCTION
Definition 4.
The probability function, also called the probability mass function,
usually denoted 𝑝𝑋 (𝑥) or 𝑝(𝑥), describes the probability at a distint
point when it is not 0. the formal definition : 𝑝𝑋 𝑥 = Pr(𝑋 = 𝑥).
Example :
Note :
For discrete random variables, the distribution and survival fuction can
be recovered as 𝐹 𝑥 = σ𝑦≤𝑥 𝑝(𝑦) and 𝑆 𝑥 = σ𝑦>𝑥 𝑝(𝑦).
THE HAZARD RATE
Defintion 5.
The hazard rate, also known as the force of mortality and the failure
rate, and usually denoted by ℎ𝑋 𝑥 or ℎ(𝑥), is the artio of the density
and survival functions when the density function is defined. That is,
𝑓𝑋 𝑥
ℎ𝑋 𝑥 = .
𝑆𝑋 𝑥
For example : For our models :
THE HAZARD RATE
Figure 1. Hazard rate function for model 1.
Figure 2. Hazard rate function for model 2.

EXERCISES
1. A random variable 𝑋 has density function 𝑓 𝑥 = 4𝑥 1 + 𝑥 2 −3 ,
𝑥 > 0. Determine the mode of 𝑋.
2. A nonnegative random variable has a hazard rate function of
ℎ 𝑥 = 𝐴 + 𝑒 2𝑥 , 𝑥 ≥ 0. You are also given 𝑆 0.4 = 0.5. determine
the value of 𝐴.
3. 𝑋 has a Pareto distribution with parameters 𝛼 = 2 and 𝜃 = 10,000.
𝑌 has a Burr distribution with parameters 𝛼 = 2. 𝛾 = 2, and 𝜃 =
20,000. Let 𝜏 be the ratio of Pr(𝑋 > 𝑑) to Pr(𝑌 > 𝑑) . Determine
lim 𝑟.
𝑑→∞
BASIC DISTRIBUTIONAL QUANTITY
(MOMENT)
Definiton :
The kth raw moment of a random variable is the expected (average)
value of the kth power of the variable, provided it exists. It is denoted
by Ε(𝑋 𝑘 ) or by 𝜇𝑘′ . The first raw moment is called the mean of the
random variable and is usually denoted by 𝜇.
The formula for the kth raw moment is :
(MOMENT)
Example : determine the first two raw moments for each of the five
models.
(MOMENT)
Solution :
(MOMENT)
Solution :
EMPIRICAL MODEL
Definition.
The empirical model is a discete distribution based on sample of size 𝑛
which assigns probability 1/𝑛 to each data point.
Example :
Consider a sample of 8 in which the observed data points were
3,5,6,6,6,7,7 and 10. The empirical model then has probability
function:
EMPIRICAL MODEL
Determine the first two raw moments for empirical model below :
Solution :
THE KTH CENTRAL MOMENT
Definition.
The kth central moment of a random variable is the expected value of
the kth power of the deviation of the variable from its mean. It is
denoted by Ε[ 𝑋 − 𝜇 𝑘 ] or by 𝜇𝑘 . The second central moment is
usually called the variance and denoted by 𝜎 2 and its square root 𝜎, is
called the standard deviation. The ratio of the standard deviation to the
mean is called the coefficient of variation. The ratio of thord central
𝜇3
moment to the cube of the standard deviation, 𝛾1 = 3 , is called
𝜎
skewness. The ratio of the fourth central moment to the fourth power
𝜇4
of the standard deviation, 𝛾2 = 4 , is called kurtosis.
𝜎
THE KTH CENTRAL MOMENT
The continous and discrete formulas for calculating central moments
are :
EXERCISES
1. A random variable has a mean and a coefficient of variation of 2.
the third raw moment is 136. determine the skewness.
2. Determine the skewness of a gamma distribution that has a
coefficient of variation of 1.
3. The severity distribution of individual claim has pdf 𝑓 𝑥 =
2.5𝑥 −3.5 , 𝑥 ≥ 1. determine the coefficient of variation.
4. Claim sizes are for 100, 200, 300, 400, or 500. the true probabilities
for these values are 0.05, 0.20, 0.50, 0.20, 0.05, respectively.
Determine the skewness and kurtosis for this distribution.
EXCESS LOSS VARIABLE
(LEFT TRUNCATED AND SHIFTED VARIABLE)
Definition :
For a given value of 𝑑 with Pr 𝑋 > 𝑑 > 0, the excess loss variable is
𝑌 = 𝑋 − 𝑑, given that 𝑋 > 𝑑. Its expected value,
𝑒𝑿 𝑑 = 𝑒 𝑑 = Ε 𝑌 = 𝐸(𝑋 − 𝑑|𝑋 > 𝑑)
Is called the mean excess loss function.
Other names for this expectation are mean residual life function and
complete expectation of life.
The 𝑘th moment of the excess loss variable is determined
from
∞
𝑘 ‫𝑥 𝑑׬‬−𝑑 𝑘 𝑓 𝑥 𝑑𝑥
1. 𝑒𝑋 𝑑 =
1−𝐹(𝑑)
, if the variable is continous.
𝑘
σ𝑥 >𝑑 𝑥𝑗 −𝑑 𝑝 𝑥𝑗
𝑗
2. 𝑒𝑋𝑘 𝑑 = 1−𝐹(𝑑)
, if the variable is discrete.
∞ ∞
‫𝑥( 𝑑׬‬ − 𝑑)𝑓 𝑥 𝑑𝑥 ‫)𝑑(𝑆 𝑑׬‬
𝑒𝑋 𝑑 = = .
1 − 𝐹(𝑑) 𝑆(𝑑)
LEFT CENSORED AND SHIFTED VARIABLE
Definition :
The left censored and shifted variable is :
The moments can be calculated from

𝑘 ∞
1. Ε[ 𝑋 − 𝑑 +] = ‫𝑑׬‬ 𝑥 − 𝑑 𝑘 𝑓 𝑥 𝑑𝑥. If the variable is continous.
𝑘 𝑘
2. Ε[ 𝑋 − 𝑑 +] = σ𝑥𝑗 >𝑑 𝑥𝑗 − 𝑑 𝑝 𝑥𝑗 . If the variable is discrete.
LEFT CENSORED AND SHIFTED VARIABLE
Note that :
𝑘
Ε 𝑋−𝑑 + = 𝑒 𝑘 𝑑 [1 − 𝐹(𝑑)]
Example :
Construct the graph to illustrate the difference between the excess loss
variable and the left censored and shifted variable.
Excess loss variable Left censored and shifted variable

THE LIMITED LOSS VARIABLE
Definition :
The limited loss variable is
Its expected value, Ε[𝑋 ∧ 𝑢], is called the limited expected value.
The 𝑘th moment of the limited loss variable are
1. If the random variable is continuous :
𝑢
Ε 𝑋∧𝑢 𝑘 = න 𝑥 𝑘 𝑓 𝑥 𝑑𝑥 + 𝑢𝑘 [1 − 𝐹(𝑢)]
−∞
2. If the random variable is discrete :
Ε 𝑋∧𝑢 𝑘 = ෍ 𝑥𝑗𝑘 𝑝 𝑥𝑗 + 𝑢𝑘 [1 − 𝐹(𝑢)]

𝑥𝑗 ≤𝑢
Another interesting formula :
0 𝑢
𝑘
Ε 𝑋∧𝑢 = − න 𝑘𝑥 𝑘−1 𝐹 𝑥 𝑑𝑥 + න 𝑘𝑥 𝑘−1 𝑆(𝑥) 𝑑𝑥
−∞ 0
For 𝑘 = 1, we have
0 𝑢
Ε 𝑋∧𝑢 𝑘 = − න 𝐹 𝑥 𝑑𝑥 + න 𝑆(𝑥) 𝑑𝑥
−∞ 0
EXERCISES
1. For two random variables, 𝑋 and 𝑌, 𝑒𝑌 30 = 𝑒𝑋 30 + 4. let 𝑋
have a uniform distribution on the interval from 0 to 100 and let 𝑌
have a uniform distribution on the interval from 0 to 𝑤. Determine
𝑤.
2. A random variable has density function 𝑓 𝑥 = 𝜆−1 𝑒 −𝑥/𝜆 , 𝑥, 𝜆 > 0.
determine 𝑒(𝜆), the mean residual life function evaluated at 𝜆.
3. Show that the following relationship holds :
Ε 𝑋 = 𝑒 𝑑 𝑆 𝑑 + Ε[𝑋 ∧ 𝑑].
EXERCISES
Loses have a Pareto distribution with 𝛼 = 0.5 and 𝜃 = 10,000 .
determine the mean residual life at 10,000.

General Insurance 2

Uploaded by

Copyright:

Available Formats

General Insurance 2

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

General Insurance 2

Uploaded by

Copyright:

Available Formats

GENERAL INSURANCE

Figure 2. Survival function for model 2.

Figure 2. Hazard rate function for model 2.

The moments can be calculated from

Excess loss variable Left censored and shifted variable

Ε 𝑋∧𝑢 𝑘 = ෍ 𝑥𝑗𝑘 𝑝 𝑥𝑗 + 𝑢𝑘 [1 − 𝐹(𝑢)]

You might also like