Exam Strategy:

1 Maximize the number of questions

answered correctly.
ACTEX Seminar
Exam P 2 Do the easier problems first.

Written & Presented by

Matt Hassett, ASA, PhD
1
3 Dont spend too much time on one
question.
4

Remember: Points to Remember:

1 This is a review seminar.

It assumes that you have already studied
y Ablackboard;
TV screen holds less content than a
use your handout pages for
probability. overview.

2 This is an actuarial exam seminar.

We will focus more on problem solving
y Algebra and calculus skills are assumed
and required.
than proofs.
y Expect multiple skill problems.
3 This is an eight hour seminar.
You may want to study more material. 2
y Calculators? 5

Other Study Materials: Probability Rules:

y Probability for Risk Management

(Text and Solutions manual)
Matt Hassett & Donald Stewart y Negation Rule:
ACTEX Publications P(E) = P(~ E) = 1- P(E)

y ACTEX Study Guide

(SOA Exam P/CAS Exam 1)
y Disjunction Rule:
P(A B) = P(A) + P(B) - P(A B)
Sam Broverman
ACTEX Publications
3 6

Copyright ACTEX 2006

 Probability Rules, cont.: Venn Diagrams Can Help:
You are given:
y Definition: P ( A B ) = 0.7 and P ( A B ) = 0.9.
Two events A and B are called
mutually exclusive if A B = Determine P[ A].

y Addition Rule
for Mutually Exclusive Events:
(A) 0.2 (B) 0.3 (C) 0.4 (D) 0.6 (E) 0.8

If AB=, P(AB) = P(A) + P(B)

7 10

Exercise: Venn Diagrams Can Help:

The probability that a visit to a primary A B Unshaded region
care physicians (PCP) office results in neither
lab work nor referral to a specialist is 35% . Of P ( A B ) = 0.9.
0.10
those coming to a PCPs office, 30% are Area of the shaded
referred to specialists and 40% require lab region must be 0.10
work.
The total area of the two circles represents:
Determine the probability that a visit to a
PCPs office results in both lab work and P ( A B ) = 0.7
referral to a specialist. Subtracting the area of the shaded region:
(A) 0.05 (B) 0.12 (C) 0.18 (D) 0.25 (E) 0.35 P(A) = 0.7 0.1 = 0.6 Answer D
8 11

Solution: A More Complicated Venn Diagram:

Let L be lab work and S be a visit to a specialist. An insurance company has 10,000
policyholders. Each policyholder is classified as
P ( L S ) = 0.35 = 1 P ( L S ) young/old; male/female; and married/single.
P ( L S ) = 0.65 Of these, 3,000 are young, 4,600 are
male, and 7,000 are married. They can also be
P(S) = 0.30 and P(L) = 0.40 classified as 1,320 young males, 3,010 married
P ( L S ) = 0.65 = P(L) + P(S) P ( L S ) males, and 1,400 young married persons. 600
are young married males.
= 0.40 + 0.30 P ( L S )
How many policyholders are young, female,
P ( L S ) = 0.05 Answer A and single?
9
(A) 280 (B) 423 (C) 486 (D) 880 (E) 896 12

Copyright ACTEX 2006

 A More Complicated Venn Diagram: Trickier Problem, cont.:

M A B
Y=3,000
720 0 0 Find
y
600 P ( A B C )
0
800
x z = 1 ( x + y + z )
0
C
H
Y = young
M = male This is a linear system for x, y, z.
H = married
13 16

A More Complicated Venn Diagram: Trickier Problem, cont.:

A B 1
M P(A) = x + y + 0 =
Y=3,000 0 y 0 4
880 720 1
0 P(B) = 0 + y + z =
600 x z
3
800 0 5
C P(C) = x + 0 + z =
12
Solution:
H x = 2/12, y = 1/12, z = 3/12
Y = young 6 1
M = male 3, 000 720 600 800 = 880 P ( A B C ) = 1 ( x + y + z ) = 1 =
12 2
H = married Answer D Answer C
14 17

Some Problems are Trickier: More Probability Rules:

An insurer offers a health plan to the
employees of a company. As part of this plan,
y Conditional probability by
counting for equally likely outcomes
each employee may choose exactly two of the n ( A B)
supplementary coverages A, B, and C, or may P(A|B) =
n (B )
choose no supplementary coverage.
The proportions of the employees that y Definition:
For any two events A and B, the
choose coverages A, B, and C are 1/4, 1/3, and conditional probability of A given B
5/12, respectively. is defined by
Determine the probability that a randomly P ( A B)
chosen employee will choose no coverage. P(A|B) =
(A) 0 (B) 47/144 (C) 1/2 (D) 97/144 (E) 7/9 15 P (B ) 18

Copyright ACTEX 2006

 More Probability Rules: Solution:

A: 312 H: 210
Multiplication Rule for Probability
P ( A B ) = P(A|B) P ( B ) 937
A = At least one parent with heart disease
H = Died of causes related to heart disease
n ( H A )
Find P ( H | A ) =
n ( A)
19 22

Exercise: Solution:

A researcher examines the medical records A: 312 H: 210

of 937 men and finds that 210 of the men died 102 108
from causes related to heart disease. 937
A = At least one parent with heart disease
312 of the 937 men had at least one H = Died of causes related to heart disease
parent who suffered from heart disease, and, of
n ( A ) = 312
these 312 men, 102 died from causes related
n ( ~ A ) = 937 312 = 625
to heart disease.
n ( A H ) = 102
20
n(H A) = n(H) 102 = 108 23

Exercise, cont.: Solution:

Find the probability that a man randomly A: 312 H: 210

selected from this group died of causes related 102 108
to heart disease, given that neither of his 937
parents suffered from heart disease.
A = At least one parent with heart disease
(A) 0.115 H = Died of causes related to heart disease
(B) 0.173 n ( H A ) 108
(C) 0.224 P ( H | A ) = = = 0.173
n ( A) 625
(D) 0.327
(E) 0.514 21
Answer B 24

Copyright ACTEX 2006

 A Harder Conditional Problem: Harder Problem Solution:
An actuary is studying the prevalence of We want to find
three health risk factors, denoted by A, B, and C, P( A B C)
P( A B C | A) =
within a population of women. For each of the P ( A)
three factors, the probability is 0.1 that a
woman in the population has only this risk factor A B P[ (A B C)]
=
(and no others). For any two of the three factors, P ( A)
the probability is 0.12 that she has exactly these
x
two risk factors (but not the other).
The probability that a woman has all three But , P ( A B C ) = x
is not given.
risk factors, given that she has A
C
and B, is 1/3.
25 28

A Harder Conditional Problem, cont.: Harder Problem Solution, cont.:

What is the probability that a woman has Fill in 0.12 in each of the areas representing
none of the three risk factors, given that she exactly two risk factors , and fill in 0.10 in each
does not have risk factor A? of the areas representing exactly one risk factor.

(A) 0.280 A B
(B) 0.311
(C) 0.467 .10 .12 .10
(D) 0.484 x
(E) 0.700 .12 .12
.10
26 C 29

DeMorgans Laws: Harder Problem Solution, cont.:

Probability of a woman having all three risk
factors given that she has A and B is 1/3.
A B = ( A B ) P ( A B C| A B)
A B = ( A B ) A B P (A B C) 1
= =
.12 P ( A B) 3
.10 .10
x P (A B C) = x
.12 .12 P ( A B ) = x + 0.12
x 1
.10 = x = 0.06
27 C x + 0.12 3 30

Copyright ACTEX 2006

 Harder Problem Solution, cont.: Exercise:
P ( A ) = 0.06 + 0.12 + 0.12 + 0.10 An actuary studying insurance preferences makes
P ( A ) = 0.60 the following conclusions:
= 0.40
(i) A car owner is twice as likely to purchase
P ( A B C ) = 0.06 + 3 ( 0.12 ) + 3 ( 0.10 ) = 0.72 collision coverage as disability coverage.
A B (ii) The event that a car owner purchases
P ( A B C )
.12 collision coverage is independent of the event
.10 .10
= 1 0.72 that he or she purchases disability coverage.
.06
.12 .12 = 0.28 (iii) The probability that a car owner
purchases both collision and disability
.10 coverages is 0.15 .
C 31 34

Harder Problem Solution, cont.: Exercise, cont.:

What is the probability that an automobile
P[ (A B C)]
P( A B C | A) = owner purchases neither collision nor disability
P ( A) coverage?
0.28
A B = (A) 0.18
0.60 (B) 0.33
.10 .12 .10 = 0.467 (C) 0.48
.06 (D) 0.67
.12 .12 (E) 0.82
.10 Answer C
C 32 35

More Probability Rules: Solution:

Let C be collision insurance and D be disability
y Definition: insurance.
Two events A and B, are independent if We need to find P ( C D ) = 1 P ( C D ) .
P(A |B) = P(A)
i) P(C) = 2 P(D)
ii) P ( C D ) = P(C) P(D)
yMultiplication Rule for
Independent Events iii) P(C D) = 0.15
If A and B, are independent,
P ( A B ) = P ( A ) P (B )
33 36

Copyright ACTEX 2006

 Solution, cont.: Solution, cont.:
0.15 = P ( C D ) = P(C) P(D) = 2P(D) 2
T 0.0095 = P ( D T )
.95
P(D) = 0.075 P(D) = 0.075
2
D ~T
.01 .05
P(C) = 2 P(D) = 2 0.075
.005 T 0.00495 = P ( ~ D T )
P(C D) = P(C) + P(D) P(C D) .99 ~D
.995 ~T
= 2 0.075 + 0.075 0.15
= 0.67 P (D T ) 0.0095
P ( D |T ) = = = 0.657
P ( C D ) = 1 P ( C D ) = 1 0.67 = 0.33 P (T ) 0.0095 + 0.00495
Answer B 37
Answer B 40

Bayes Theorem -- Simplify with Trees: Probability Rules:

A blood test indicates the presence of a
particular disease 95% of the time when the
disease is actually present. The same test Law of Total Probability:
indicates the presence of the disease 0.5% of Let E be an event. If A1, A2 , An
the time when the disease is not present. 1% of partition the sample space, then
the population actually has the disease. P ( E ) = P ( A1 E ) + P ( A2 E ) + ... +P ( An E ) .
Calculate the probability that a person has
the disease given that the test indicates the
presence of the disease.
(A)0.324 (B) 0.657 (C) 0.945
(D) 0.950 (E) 0.995
38 41

Solution: Theorem:
D = Person has the disease
T = Test indicates the disease Bayes Theorem:
We need to find Let E be an event. If A1, A2 , An
partition the sample space, then
P (D T ) P ( E A1 )
P ( D |T ) = P ( A1| E ) =
P (T ) P (E )
P ( A1 ) P ( A1| E )
=
P ( A1 ) P ( A1| E ) + + P ( An ) P ( An | E )

39 42

Copyright ACTEX 2006

 Exercise:
An insurance company issues life insurance
policies in three separate categories: standard,
preferred, and ultra-preferred. Of the
companys policyholders, 50% are standard,
40% are preferred, and 10% are ultra-
preferred. Each standard policyholder has
probability 0.010 of dying in the next year,
each preferred policyholder has probability
0.005 of dying in the next year, and each ultra-
preferred policyholder has probability 0.001 of
dying in the next year. 43
Exercise:
The probability that a randomly chosen
male has a circulation problem is 0.25 . Males
who have a circulation problem are twice as
likely to be smokers as those who do not have a
circulation problem.
What is the conditional probability that a
male has a circulation problem, given that he a
smoker?
(A) 1/4 (B) 1/3 (C) 2/5 (D) 1/2 (E) 2/3 46

Exercise, cont.: Solution:

A policyholder dies in the next year. C = Circulatory problem
S = Smoker
What is the probability that the deceased
policyholder was ultra preferred?
y We need to find P ( C | S ) .
(A) 0.0001 y We do not know x = P ( S |~ C ) .
(B) 0.0010
(C) 0.0071
y We do know that 2x = P ( S | C ) since
those who have a circulation problem are
(D) 0.0141 twice as likely to be smokers.
(E) 0.2817
44 47

Solution Solution, cont.:

.01 D .005
S 2x S .5x
.5 C
.005 D .002 .25
.4 P
.1 .001 D .0001 .75 x S .75x
U ~C

P (U D )
P (U | D ) = P (C S ) .5x .5
P (D ) P ( C|S ) = = = = .40
.0001 P (S) .5x + .75x 1.25
= = .0141
.0001+ .002 + .005 45
Answer D Answer C 48

Copyright ACTEX 2006

 Expected Value: Exercise, cont.:
y Definition: What percentage of the claims are within one
standard deviation of the mean claim size?
The expected value of X is defined by
E (X ) = x p ( x) (A)45%
The expected value is also referred to as the (B) 55%
mean of the random variable X and denoted (C) 68%
by Greek letter . E ( x ) = . (D) 85%
(E) 100%
y A Property of Expected Value:
E ( aX + b ) = a E ( X ) + b
49 52

Variance: Solution:
y Definition:
The variance of a random variable X is
Claim Size Probability xp(x) x2p(x)
20 0.15 3 60
V ( X ) = E ( X ) = ( x ) p ( x )
2 2
30 0.10 3 90
40 0.05 2 80
y Standard Deviation:
50 0.20 10 500
= V ( X ) . Notation: V ( X ) = 2 60 0.10 6 360

y V (X ) = E (X ) E (X ) = E (X 2 ) 2
70 0.10 7 490
2 2

80 0.30 24 1920
y V ( aX + b ) = a V ( X )
2

50
Total 1.00 55 3500
53

Exercise: Solution, cont.:

A probability distribution of claim sizes is given E ( X ) = x p ( x ) = 55
in this table:
Claim Size Probability 2 = V ( X ) = E ( X 2 ) = 2 = 3500 552 = 475
20 0.15
= 475 = 21.8
30 0.10
A value is within one standard deviation of the
40 0.05
mean if it is in the interval [ , + ] ,
50 0.20 that is, in the interval [33.2, 76.8] .
60 0.10
70 0.10 20 30 40 50 60 70 80
80 0.30 51
33.2 55 76.8 54

Copyright ACTEX 2006

 Solution, cont.: Additional Properties of V(X):
The values of x in this interval are 40, 50, 60,
and 70. y V ( X + Y ) = V ( X ) + V (Y ) + 2cov ( X , Y )
20 30 40 50 60 70 80

33.2 55 76.8 y For X, Y independent

V ( X + Y ) = V ( X ) + V (Y )
Thus, the probability of being within one
standard deviation of the mean is:
p(40) + p(50) + p(60) + p(70)
= .05 + .20 + .10 + .10 = .45
55 58

Z-Score: Exercise:
The profit for a new product is given by
Z = 3X Y 5.
Definition:
For any possible value x of a random variable, X and Y are independent random variables with
x- V(X) = 1 and V(Y) = 2.
z=
What is the variance of Z?
The z score measures the distance of x from
E ( X ) = in standard deviation units.
(A) 1 (B) 5 (C) 7 (D) 11 (E) 16

56 59

Theorem: Solution:
V (Z) = V ( 3X Y 5) = V ( 3X Y )
y Chebychevs Theorem: = V ( 3X + ( Y ) ) = V ( 3X ) + V ( Y )
For any random variable X, the probability independence

that X is within k standard deviations of the = 32 V ( X ) + ( 1) V (Y )

2

1
mean is at least 1 2 . = 9 (1) + 2 = 11 Answer D
k
Note!
Observe the wrong answer which you would obtain
1
y P( - k X + k ) 1- k 2
if you mistakenly wrote V (3X - Y ) = V (3X ) - V (Y ).
This is choice C, and is a common careless mistake.
57 60

Copyright ACTEX 2006

 Exercise: Geometric Distribution:
A recent study indicates that the annual cost of
P ( X = k ) = q k p, k = 0, 1, 2, 3,
maintaining and repairing a car in a town in
Ontario averages 200 with a variance of 260. q q
E (X ) = V (X ) =
p p2
If a tax of 20% is introduced on all items
associated with the maintenance and repair of
cars (i.e., everything is made 20% more where X= the number of failures before the
expensive), what will be the variance of the first success in a repeated series of
annual cost of maintaining and repairing a car? independent success-failure trials with
P ( Success ) = p.
(A) 208 (B) 260 (C) 270 (D) 312 (E) 374
61 64

Solution: Geometric Distribution Alternative:

Let X be the random variable for the present Here, you are looking at the number of trials
cost, and Y=1.2X the random variable for the needed to get to the first success. In this
cost after 20% inflation. We are asked to formulation, you are looking at Y = X + 1.
find V(Y). P (Y = k ) = q k 1 p , k = 1, 2, 3,
V (Y ) = V (1.2X ) 1 q
. E (Y ) = V (Y ) =
= 1.22 V (X ) p p2
= 1.44 ( 260 )
= 374.4 Answer E 62 65

Geometric Series Review: Exercise:

In modeling the number of claims filed by
y A geometric sequence is of the form
2 3
a , ar , ar , ar , ..., ar .n an individual under an automobile policy during
a three-year period, an actuary makes the
y The sum of the series for r 1 is given by: simplifying assumption that for all integers n 0,
1
1 r n+1 p n +1 = p n where p n represents the probability
a + ar + ar 2 + ... + ar n = a 5
1- r that the policyholder files n claims during the
The number r is called the ratio. period.
y If |r|<1, we can sum the infinite geometric Under this assumption, what is the
probability that a policyholder files more than
series:
1
a + ar + ar 2 + ... + ar n + ... = a one claim during the period?
1- r
63 66
(A) 0.04 (B) 0.16 (C) 0.20 (D) 0.80 (E) 0.96

Copyright ACTEX 2006

 Solution: Binomial Distribution:
We are not given p 0 . Look at the first few Given n independent, success-failure trials with
terms: 2 2 P(S) = p , P ( F ) = 1 p = q
1 1 1
p0 , p0 , p0 , p0 , n
5 5 5 y P(X = k) = p k (1 p) n-k
k
n
= p k (q) n-k , k = 0, 1, , n
k
y E ( X ) = np
V ( X ) = np (1 p ) = npq
67 70

Solution, cont.: Notation Review:

1 1 2 1 3 1
n

1 = p 0 1+ + + + ... + + ...
y n ! = n ( n 1) ... (2)1
5 5 5 5
y ( nr ) = C(n, r) = r !
P(n , r)

1 5 n!
= p0 = p0 =
1 4 r !(n r)!
1 n (n 1) (n r + 1)
5 =
4 r!
p0 =
5 10 10! 10 9
68
y 2 = 2!8! = 2 1
= 45
71

Solution, cont.: Example:

4 4 1 Guessing on a 10 question multiple choice
P ( N > 1) = 1 P ( N 1) = 1 + = .04 quiz with choices A, B, C, D, E.
5 5 5
n = 10, P ( S ) = .2 = p , q = .8
Answer A
10
P ( X = 2 ) = ( .2 ) ( .8 ) .302
2 8

Note! 2
The probability distribution has the form of a
1 E ( X ) = 10 ( .2 ) = 2
geometric distribution with q = , so it must be
4 5 V ( X ) = 10 ( .2 )( .8 ) = 1.6
true that p 0 = p = .
5 69 72

Copyright ACTEX 2006

 Exercise:
A study is being conducted in which the
health of two independent groups of ten
policyholders is being monitored over a one-
year period of time. Individual participants in the
study drop out before the end of the study with
probability 0.2 (independently of the other
participants).
What is the probability that at least 9
participants complete the study in one of the two
groups, but not in both groups?
(A) .096 (B) .192 (C) .235 (D) .376 (E).469 73
Solution:
Denote the random variables for the number
of participants completing in each group by
A and B. We need
P ( A 9 & B < 9 ) or ( B 9 & A < 9 )
= P ( A 9 & B < 9 ) + P (B 9 & A < 9 )
= P ( A 9) P (B < 9) + P (B 9) P ( A < 9)
Ind
The two groups are independent and have
identical binomial probability distributions.
74
Solution, cont.:
A is binomial with n=10 independent trials and
probability of completion p=0.8.
P ( A 9 ) = P ( A = 10 ) + P ( A = 9 )
10
= .810 + .89 ( .2 ) = .376
9 of defectives in 30 components is a binomial
P(A < 9) = 1 P ( A 9 ) = .624
P ( B 9 ) = .376 P(B < 9) = .624
P ( A 9) P (B < 9) + P (B 9) P ( A < 9)
= .376 ( .624 ) + .376 ( .624 ) = .469
75
Answer E
Copyright ACTEX 2006
Harder Bayes Thrm./Binomial Exercise:
A hospital receives 1/5 of its flu vaccine
shipments from Company X and the remainder of
its shipments from other companies. Each
shipment contains a very large number of
vaccine vials.
For Company Xs shipments, 10% of the
vials are ineffective. For every other company,
2% of the vials are ineffective. The hospital tests
30 randomly selected vials from a shipment and
finds that one vial is ineffective.
76
Bayes Thrm./Binomial Exercise, cont.:
What is the probability that this shipment
came from Company X?
(A) 0.10
(B) 0.14
(C) 0.37
(D) 0.63
(E) 0.86
77
Solution:
X = Shipment came from company X
I = Exactly 1 vial out of 30 tested is ineffective
We are asked to find P ( X | I ) .
If the shipment is from company X, the number
random variable with n=30 and p=0.1.
The probability of one defective in a batch of
30 from X is
30
P ( I | X ) = ( .1) ( .9 29 ) = .141
1
78
 Solution, cont.: Example:
X = Shipment came from company X Accidents occur at an average rate of = 2
I = Exactly 1 vial out of 30 tested is ineffective per month.
We are asked to find P ( X | I ) .
Let X = the number of accidents in a month.
If the shipment isnt from company X, the number
of defectives in 30 components is a binomial e 2 21
P ( X = 1) = .271
random variable with n=30 and p=0.02. 1!
E (X ) = V (X ) = 2
30
P ( I |~ X ) = ( .02 ) ( .9829 ) = .334
1
79 82

Solution, cont.: Exercise:

An actuary has discovered that
.141 I .0282 policyholders are three times as likely to file two
X claims as to file four claims. If the number of
.2
claims filed has a Poisson distribution, what is the
.334 variance of the number of claims filed?
.8 I .2672
~X (A) 1/ 3
(B) 1
P (X I) .0282
P ( X |I ) = = = .0955 (C) 2
P (I ) .0282 + .2672 (D) 2
Answer A 80
(E) 4 83

Poisson Distribution: Solution:

X is Poisson with mean .
P ( X = 2) = 3 P ( X = 4 )
e k
y P(X = k) =
k!
, k = 1, 2, 3, e 2 e 4
= 3
2!
4!
y E (X ) = 4 2 = 4 = 2
V (X ) =

81
Answer D 84

Copyright ACTEX 2006

 Hypergeometric Example: Previous Example, cont.:
A company has 20 male and 30 female X = number of males chosen in a sample of 5.
N = 50 n=5 r = 20
employees. 5 employees are chosen at random
for drug testing. What is the probability that 3 30 20
5- kk
males and 2 females are chosen?
P(X = k) =
50
Solution: 5
20 30
3 2 20
0.234 E(X ) = 5 = 2
50 50
5 20 20 50 - 5
V (X ) = 5 1
50 50 50 -1
85 88

Hypergeometric Probabilities: Negative Binomial Distribution:

A series of independent trials has P(S) = p on

1 A sample of size n is being taken from a

finite population of size N.
each trial.
Let X be the number of failures before success r.
r + k -1 k r
2 The population has a subgroup of size r n
that is of interest.
y P(X = k) =
r -1
q p , k = 0, 1, 2, 3,

rq rq
y E(X) = p V (X) = p
3
The random variable of interest is X, the 2
number of members of the subgroup in
the sample taken.
86
y The special case with r = 1 is the geometric
random variable. 89

Hypergeometric Probabilities, cont.: Example:

N - r r Play slot machine repeatedly with probability of

n - k k success on each independent play P ( S ) = .05 = p.
y P(X = k) = , k = 0, , n

N Find the probability of exactly 4 losses (failures)
n before the second win (success r=2).

r
y E (X ) = n
N
r r N - n
V ( X ) = n 1
N N N -1 87 90

Copyright ACTEX 2006

 Example, cont.: Solution:
Possible sequences: This is a negative binomial distribution problem.
SFFFFS Success S = month with at least one accident
FSFFFS Failure F = month with no accidents.
FFSFFS
FFFSFS Note that P(S) = p = 3/5.
FFFFSS Let X be the number of months with no accidents
Single sequence probability : .052 ( .95)
4
before the fourth month with at least one
5 accident i.e., the number of failures before the
Number of sequences: = 5
fourth success.
1
X is negative binomial with r = 4 and p=3/5.
Solution: 5 ( .05) ( .95) .0818
2 4
91 94

Exercise: Solution, cont.:

We are asked to find
A company takes out an insurance policy to P ( X 4 ) = 1 P ( X 3 )
cover accidents at its manufacturing plant. The = 1 P ( X = 0 ) +4 P ( X = 1) + P ( X = 2 ) + P ( X = 3 )
probability that one or more accidents will occur 3
P ( X = 0 ) = = 0.12960
during any given month is 3/5. The number of 5 4
accidents that occur in any given month is 4 3 2
P ( X = 1) = = .20736
independent of the number of accidents that 1 5 4 5 2
occur in all other months. 5 3 2
P ( X = 2 ) = = .20736
2 5 4 5 3
6 3 2
P ( X = 3 ) = = .16589
3 5 5
92 95

Exercise, cont.: Solution, cont.:

Calculate the probability that there will be P ( X 3 ) = .12960 + .20736 + .20736 + .16589
at least four months in which no accidents occur
before the fourth month in which at least one = .71021
accident occurs.
P ( X 4 ) = 1 P ( X 3 )
(A) 0.01 (B) 0.12 (C) 0.23 (D) 0.29 (E) 0.41 = 1 .71021 = .28979

Answer D
93 96

Copyright ACTEX 2006

 Definition of Continuous Distribution: Uniform Random Variable on [a, b]:
The probability density function of a random
variable X is a real valued function satisfying: 1
y f ( x) = b - a , a x b
1f (x) 0 for all x.
0, otherwise
2 The total area bounded by the graph of
y = f (x) and the x axis is 1.

a+b (b - a)2

f (x) dx = 1
y E(X ) = V (X ) =
2 12
3 P ( a X b ) is given by the area under
y = f (x) between x = a and x = b.
b
P(a X b) = f (x) dx
a 97 100

Continuous Distribution Properties: Exponential Distribution:

yCumulative Distribution Function F(x)
x
y Random variable T, parameter .
F(x) = P(X x) = f (u) du
T is often used to model waiting time, = rate.
y Expected Value

E(X ) = x f ( x ) dx
y f(t) = e
- t

, F(t ) = 1- e - t for t 0
y Expected value of a function of a
continuous random variable

E g ( X ) = g ( x ) f ( x ) dx 1 1

y E(T ) = V (T ) =
2
y Mean of Y = aX + b
E ( aX + b ) = a E ( X ) + b 98 101

Continuous Distribution Properties: Example:

y Variance
Waiting time for next accident.
V (X ) = E[(X - )2 ] = (x - )2 f (x) dx = 2 accidents per month on average.

P ( 0 T 1) = F(1) = 1 e 2 .865
y Alternate Form
of Variance Calculation 1 1
V (X ) = E(X 2 ) [ E(X )] = E(X 2 ) 2
2
E (T ) = V (T ) =
2 4

y Variance of Y = aX + b
Exponential waiting time
V (aX + b) = a V (X ) * *
2
Poisson number of events
99 102

Copyright ACTEX 2006

 Useful Exponential Facts: Solution:
xn Recall, the mean of the exponential is = 1/ .
y lim x e = lim = 0, for a > 0.
n - ax

x x e
ax Thus if you are given the mean (as in this
problem), you know that 1/ = .
n! G: Waiting time for 1st accident for good driver
y 0
x n e - ax dx =
a n +1
,
B: Waiting time for 1st accident for bad driver
for a > 0, and n a positive integer.
x
1
G: G = FG (x) = 1 e 6
6
1
x
B: B = FB (x) = 1 e 3
103 3 106

Exercise: Solution, cont.:

The waiting time for the first claim from a Find P ( G 3 & B 2 ) .
good driver and the waiting time for the claim Note that G and B are independent.
from a bad driver are independent and follow
exponential distributions with 6 years and 3 P ( G 3 & B 2) = P ( G 3) P (B 2)
years, respectively. = FG ( 3 ) FB ( 2 )
What is the probability that the first claim

3

2
from a good driver will be filed within 3 years = 1 e 6 1 e 3
and the first claim from a bad driver will be
filed within 2 years?
2

1

7
= 1 e 3
e 2 +e 6
104
Answer C 107

Exercise: Exercise:
The number of days that elapse between
1
2 1 7

(A) 1 e e + e the beginning of a calendar year and the
3 2 6
18 moment a high-risk driver is involved in an
7
1 6 accident is exponentially distributed. An
(B) e
18 insurance company expects that 30% of high-risk

2

1

7
drivers will be involved in an accident during the
(C) 1 e 3
e 2
+e 6
first 50 days of a calendar year.

2

1

1 What portion of high-risk drivers are
(D) 1 e 3
e 2
+e 3
expected to be involved in an accident during
1 1 1 1
2 7
the first 80 days of a calendar year ?
(E) 1 e 3 e 2 + e 6 105
(A) 0.15 (B) 0.34 (C) 0.43 (D) 0.57 (E) 0.66 108
3 6 18

Copyright ACTEX 2006

 Solution: Exercise:
T: time in days until the first accident for a An insurance policy reimburses dental
high risk driver expense, X, up to a maximum benefit of 250.
To find: P(T 80) = F(80). The probability density function for X is:
ce 0.004 x for x 0
f ( x) =
t
We know F(t ) = 1 e , but we dont know . 0 otherwise
where c is constant.
Use the given probability for the first 50 days
to find it.
Calculate the median benefit for this policy.

109
(A) 161 (B) 165 (C) 173 (D) 182 (E) 250 112

Solution, cont.: Solution:

P(T 50) = F(50) You can see by direct examination that X must
0.004 x
= 1 e 50 be exponential with c = .004, since .004e
is the density function for the exponential with
= 0.30 = .004.
ln ( 0.7 ) (Some of our students integrated the density
=
50 function and set the total area under the curve
Now we have and can finish the problem. equal to 1, but that takes extra time.)
P(T 80) = F(80) = 1 e 80 = .4348

Answer B 110 113

Definitions: Solution, cont.:

y The mode of a continuous random variable is Original expense X: cumulative distribution
the value of x for which the density function F(x) = 1 e .004 x .
f(x) is a maximum. Thus the median m for X is obtained by solving
y The median m of a continuous random variable the equation
X is defined by F(m) = P(X m) = 0.50. F(m) = 0.50 = 1 e .004 m 0.50 = e .004 m
ln ( .50 )
yLet X be a continuous random variable and
0 p 1. The 100p percentile of X is the
th
m=
.004
= 173.3
number x p defined by F(x p ) = p. Actual benefit capped at 250. Since 173.3 is
less than 250, 50% of the benefits paid are still
Note that the 50th percentile is the median. less than 173.3 and 50% are greater.
111
Answer C 114

Copyright ACTEX 2006

 Normal Random Variable: Standard Normal Example:
X normal, = 500, = 100
y = E ( X ) and 2 = V ( X )
P ( 600 X 750 )
( x - ) 2 600 500 650 500
1 = P Z
y f ( x) =
2
e 2 2
100 100
= P (1 Z 1.5)
= .9332 .8413
= .0919
115 118

Transformation to Standard Normal: Central Limit Theorem:

Transform any normal random variable X with
mean and standard deviation into a
standard normal random variable Z with mean
0 and standard deviation 1.
X 1
Z= = X
Then probabilities can be calculated using the
standard normal probability tables for Z.
116
Let X1, X 2 , , X n be independent random
variables all of which have the same probability
distribution and thus the same mean and
variance 2 . If n is large, the sum
S = X1 + X 2 + + X n
will be approximately normal with mean n
and variance n .
2
119

Normal Distribution Table: Exercise:

An insurance company issues 1250 vision
care insurance policies. The number of claims
filed by a policyholder under a vision care
insurance policy during one year is a Poisson
random variable with mean 2. Assume the
numbers of claims filed by distinct policyholders
are independent of one another.
What is the approximate probability that
there is a total of between 2450 and 2600
claims during a one-year period?
117
(A) 0.68 (B) 0.82 (C) 0.87 (D) 0.95 (E) 1.00 120

Copyright ACTEX 2006

 Solution: Example:
X i : number of claims on policy i, i=1, , 1250. X: normal random variable with mean and
(Poisson) and standard deviation .
X i : iid with mean = 2 and variance 2 = 2. Find x p the 100pth percentile of X using the
100pth percentile of Z.
The total number of claims is xp
zp = x p = + z p
S = X1 + + X1250
By the central limit theorem, S is approximately For example, if X is a standard test score
normal with E(S) = s = 1250(2) = 2500 random variable with mean = 500 and
standard deviation = 100 then the 99th
V (S) = s 2 = 1250(2) = 2500
percentile of X is
S = 2500 = 50 x .99 = + z .99 = 500 + 2.326 (100 ) = 732.6
121 124

Solution, cont.: Exercise:

Thus A charity receives 2025 contributions.
Contributions are assumed to be independent
P ( 2450 S 2600 )
and identically distributed with mean 3125
2450 2500 2600 2500 and standard deviation 250.
= P Z
50 50 Calculate the approximate 90th
= P ( 1 Z 2 ) percentile for the distribution of the total
contributions received.
= .9772 .1581
(A) 6,328,000 (B) 6,338,000
= .8191
(C) 6,343,000 (D) 6,784,000
Answer B (E) 6,977,000
122 125

Normal Distribution Percentiles: Solution:

X i : number of contributions i, i=1, , 2025.
y The percentiles of the standard normal can X i : iid with mean = 3125 and variance
be determined from the tables. For example,
2 = ( 250 ) .
2

P(Z 1.96) = .975 The total contribution is

Thus the 97.5 percentile of the Z distribution S = X1 + + X 2025
is 1.96. By the central limit theorem, S is approximately
normal with
y Commonly
Z
used percentiles of Z:
0.842 1.036 1.282 1.645 1.960 2.326 2.576
E(S) = s = 3125(2025) = 6, 328,125
P(Z<z) 0.800 0.850 0.900 0.950 0.975 0.990 0.995 V (S) = s 2 = 250(2025) = 126, 562, 500
123
S = 126, 562, 500 = 11, 250 126

Copyright ACTEX 2006

 Solution, cont.: Corollary applied to sample mean X :
Since z .90 = 1.282, the 90th percentile of S is Let X1, X 2 , , X n be iid normal random
s .90 = 6, 328,125 + 1.282 (11, 250 ) variables with mean and variance 2 .
The sample mean is defined to be
= 6, 342, 547.5 S X1 + ... + X n
Answer C X= =
n n
For any n, the sample mean X will be normal
2
with mean and variance .
n

127 130

Theorem:
If X1, X 2 , , X n are independent normal
random variables with respective means
1, 2 , , n and respective variances
12 , 22 , , n2 , then X1 + X 2 + + X n is normal
with mean 1 + 2 + + n and variance
12 + 22 + + n2 .
Note that this shows that you dont need large n
(as required by the Central Limit Theorem) to
have a normal sum.
128
Exercise:
Claims filed under auto insurance policies
follow a normal distribution with mean 19,400
and standard deviation 5,000.
What is the probability that the average
of 25 randomly selected claims exceeds
20,000?
(A) 0.01 (B) 0.15 (C) 0.27 (D) 0.33 (E) 0.45
131

Corollary: Solution:
Let X1, X 2 , , X n be independent normal X i : claim amount on policy i, i=1, , 25.
random variables all of which have the same X i : iid with = 19, 400 and variance 2 = 5000 2.
probability distribution and thus the same The average of 25 randomly selected claims is
mean and variance .
2
S X1 + ... + X 25
X= =
25 25
For any n, the sum S = X1 + X 2 + + X n
will be normal with mean n and variance n 2 . E(X ) = = 19, 400
2 5000 2
V (X ) = = = 1000 2
25 25
129 X = 10002 = 1000 132

Copyright ACTEX 2006

 Solution, cont.: Solution:
20, 000 19, 400 X i : lifetime of light bulb i, i=1, , n.
P ( 20, 000 < X ) = P < Z X i : iid with = 3 and variance 2 = 1.
1, 000
Total lifetime of the succession of n bulbs is
= P ( .6 < Z )
S = X1 + X 2 + + X n
= .2743 E(S) = S = 3n
V (S) = S2 = n (1) = n
Answer C
S = n
The succession of light bulbs produces light for
at least 40 months with probability at least
133 136
0.9772 .

Exercise: Solution, cont.:

S 3n 40 3n
A company manufactures a brand of .9772 = P ( S 40 ) = P
light bulb with a lifetime in months that is n n
normally distributed with mean 3 and 40 3n
variance 1. A consumer buys a number of = PZ
n
these bulbs with the intention of replacing
them successively as they burn out. The light Z tables: P ( Z 2) = .9772.
bulbs have independent lifetimes. 40 3n
= 2 3n 2 n 40 = 0
n
Make the substitution x = n .
3x 2 2x 40 = 0 x = n = 4 n = 16
134
Answer B 137

Exercise:
What is the smallest number of bulbs to
be purchased so that the succession of light
bulbs produces light for at least 40 months
with probability at least 0.9772?
(A) 14 (B) 16 (C) 20 (D) 40 (E) 55
Definition:
The pure premium for an insurance is the
expected value of the amount paid on the
insurance. The amount paid is usually a function
of a random variable g(X), so to find pure
premiums we use the theorem
E g ( x ) = g ( x ) f ( x ) dx

135 138

Copyright ACTEX 2006

 Insurance with a cap or policy limit: Solution:
An insurance policy reimburses a loss up to a X: repair cost; Y: amount paid by insurance.
benefit limit of 10 . The policyholders loss, Y, Find the standard deviation Y = V (Y ) .
follows a distribution with density function:
2 0, 0 < x 250
y >1 Y=
f (y ) = y 3 x 250, 250 < x
0 otherwise

What is the expected value of the benefit paid Density function of X is f(x) = 1/1500 on the
under the insurance policy? interval (0, 1500).
(A) 1.0 (B) 1.3 (C) 1.8 (D) 1.9 (E) 2.0
139 142

Solution: Solution, cont.:

Let B=the random variable for the benefit paid.
1 < y < 10 E (Y )
y ,
B=
10 y 1 1
dx + 250 ( x 250 )
250 1500
10, = 0 dx
10 2 2 0
1500 1500
E ( B ) = y 3 dy + 10 3 dt
2 1500
1
y 10
y ( x 250 )
10 =
= 2y 1 10y 2 3000
1 10 250

1 1 = 520.833
= 2 1 10 0
10 100
= 1.9 Answer D 140 143

Insurance with a deductible: Solution, cont.:

The owner of an automobile insures it
against damage by purchasing an insurance E (Y 2 )
policy with a deductible of 250 . In the event 1 2 1
dx + 250 ( x 250 )
250 1500
that the automobile is damaged, repair costs = 0 dx
0
1500 1500
can be modeled by a uniform random
3 1500
variable on the interval (0, 1500) .
=
( x 250 )
Determine the standard deviation of the 4500
insurance payment in the event that the 250

automobile is damaged. = 434, 027.778

(A) 361 (B) 403 (C) 433 (D) 464 (E) 521
141 144

Copyright ACTEX 2006

 Solution, cont.: Solution, cont.:

V (Y ) = E (Y 2 ) E (Y ) 2 2.5 ( 0.6 ) 2.5 ( 0.6 )

2 2.5 2.5

E (Y ) = x dx + 2 dx
.6 x 3.5 2 x 3.5
= 434, 027.778 520.8332

= 162, 760.764
2
2.5 ( 0.6 ) 5 ( 0.6 )
2.5 2.5
x 1.5 x 2.5
= +
1.5 2.5
.6 2
Y = V (Y ) = .83568 + .09859
= 162, 760.76 = .93427
= 403.436
Answer B 145
Answer C 148

Exercise:
A manufacturers annual losses follow a
distribution with density function
2.5 ( 0.6 ) 2.5
for x > 0.6
f ( x) = x 3.5
0 otherwise
To cover its losses, the manufacturer
purchases an insurance policy with an annual
deductible of 2.
What is the mean of the manufacturers
annual losses not paid by the insurance policy?
146
(A) 0.84 (B) 0.88 (C) 0.93 (D) 0.95 (E) 1.00
Exercise:
An insurance policy is written to cover a
loss, X, where X has a uniform distribution on
[0, 1000].
At what level must a deductible be set in
order for the expected payment to be 25% of
what it would be with no deductible?
(A) 250 (B) 375 (C) 500 (D) 625 (E) 750
149

Solution: Solution:
X: actual cost; Y: part of loss not paid by policy. d: unknown deductible;
Find E(Y). Y: amount paid by insurance.
0, x<d
Y=
Since there is a deductible of 2, x d , x d
1
Density function for X: f (x) = , 0 x 1000
x , .6 < x < 2 1000
Y= d
2, x 2 1 1
E (Y ) = 0 dx + ( x d )
1000
dx
0
1000 d
1000
2 1000
(x d) (1000 d )
2

= =
147 2000 2000 150

Copyright ACTEX 2006

 Solution, cont.: Density Function When Inverse Exists:
Find d such that E (Y ) = .25 E ( X ) . Case 1. g(x) is strictly increasing on the
For the uniform X on [0, 1000], E ( X ) = 500 sample space for X.
and .25E ( X ) = 125. Let h(y) be the inverse function of g(x). The
function h(y) will also be strictly increasing. In
E (Y ) = .25E ( X ) this case, we can find FY(y) as follows:
(1000 d )
2 FY ( y ) = P(Y y ) = P(g(X ) y )
= 125
2000 = P h ( g ( X ) ) h(y )
(1000 d )
2
= 250, 000 d = 500 = P ( X h(y ) )
Answer C 151 = FX ( h(y ) ) 154

Finding the Density Function for Y=g(x):
Example:
Cost, X, is exponential with = .01. After
inflation of 5%, the new cost is Y = 1.05X .
Find FY (y ).
Note that FX ( x ) = 1 e .01x .
FY ( y ) = P (Y y ) = P (1.05X Y )
Density Function When Inverse Exists:
Case 2. g(x) is strictly decreasing on the
sample space for X.
Let h(y) be the inverse function of g(x). The
function h(y) will also be strictly decreasing. In
this case, we can find FY(y) as follows:
FY ( y ) = P(Y y ) = P(g(X ) y )

Y Y = P h ( g ( X ) ) h(y )
= PX = FX
1.05 1.05 = P ( X h(y ) )
y
= 1 e
.01
1.05 152 = S X ( h(y ) ) 155

Example, cont.: Density Function When Inverse Exists:

.01
Y is exponential with = . We can find the density function fY(y) by
1.05 differentiating FY(y).

Density function: fY ( y ) = FY ( y ) The final result can be written in the same way
for both cases:
Useful notation: S(x) = P ( X > x ) = 1 F ( x ) fY ( y ) = fX ( h(y ) ) h ( y )

153 156

Copyright ACTEX 2006

 Exercise: Exercise, cont.:
The time, T, that a manufacturing system is v /10,000
10, 408
(A) 10, 000e
out of operation has cumulative distribution 425
function
2 2 (B) 25e v /10,000 0.04
1 for t > 2
F(t ) = t v 10, 408
0 (C)
otherwise 10,833 10, 408
The resulting cost to the company is Y = T 2 . (D) 25
Determine the density function of Y, for y > 4. v
(A) 4 (B) 8 (C) 8 (D) 16 (E) 1024 v
y2 y 3/2 y3 y y5
(E) 25 ln .04
157
10, 000 160

Solution: Solution:
First find the cumulative distribution function for Y: xa
Uniform distribution fact to use here: F(x) = .
(
FY ( y ) = P (Y y ) = P ( T 2 y ) = P T y ) R is uniform on (0.04, 0.08)
ba

2
2 r .04
= FT ( y ) = 1 = 1
4
y
FR ( r ) =
.04
, for 0.04 r 0.08.
y
Then the density function for Y is: Find the cumulative distribution function for V.
d d 4 4
fY (y ) = FY ( y ) = 1 =
dy dy y y 2
Answer A 158 161

Exercise: Solution, cont.:

An investment account earns an annual F ( v ) = P (V v ) = P (10, 000e R v )
interest rate R that follows a uniform distribution
on the interval (0.04, 0.08). The value of a v
= P R ln
10,000 initial investment in this account after one 10, 000
year is given by V = 10, 000 e R . v
Determine the cumulative distribution v
ln .04
= FR ln 10, 000
function, F(v), of V for values of v that satisfy =
0<F(v)<1. 10, 000 .04
v
= 25 ln .04 .
159 10, 000 Answer E
162

Copyright ACTEX 2006

 Independent Random Variable Results: Exponential Random Variable Results:
General results for the minimum or maximum Minimum of independent exponential random
of two independent random variables: variables:
Recall that the survival function of a random X and Y with parameters and .
variable X is S X (t ) = P(X > t ) = 1 FX ( t ) . S Min (t ) = S X ( t ) SY ( t ) = e t e t = e
( + ))t

Min=min(X, Y) is exponential with parameter

Recall that for X exponential we have + .
F ( x ) = 1 e x and S ( x ) = e x .

163 166

Independent Random Variable Results: Exercise:

X and Y independent random variables. In a small metropolitan area, annual
losses due to storm, fire, and theft are
Find survival function for Min=min(X, Y): assumed to be independent, exponentially
distributed random variables with respective
S Min (t ) = P(min(X , Y ) > t ) means 1.0, 1.5, and 2.4 .
= P(X > t & Y > t ) = P(X > t ) P(Y > t ) Determine the probability that the
independence
maximum of these losses exceeds 3.
= S X ( t ) SY ( t )
(A) 0.002 (B) 0.050
(C) 0.159 (D) 0.287
164
(E) 0.414 167

Independent Random Variable Results: Solution:

X and Y independent random variables. X1, X 2 , X 3 : losses due to storm, fire, and theft,
respectively.
Find cumulative distribution for Max=max(X, Y): Find P [ Max > 3] , where Max = max ( X1, X 2 , X 3 ) :
FMax (t ) = P(max(X , Y ) t ) FMax (t ) = FX1 ( t ) FX2 ( t ) FX3 ( t )
= P(X t & Y t ) =
independence
P(X t ) P(Y t ) = (1 e x )(1 e x /1.5 )(1 e x / 2.4 )
= FX ( t ) FY ( t ) P(Max 3) = FMax (3)
= (1 e 3 )(1 e 3/1.5 )(1 e 3/ 2.4 ) = .586
P [ Max > 3] = 1 .586 = .414
165
Answer E 168

Copyright ACTEX 2006

 Moments of a Random Variable: MGF Useful Properties:

y Definition: MaX +b ( t ) = e tb MX ( at )
The nth moment of X is E ( X n ) .
If a random variable X has the moment
y Moment Generating Function: generating function of a known distribution,
then X has that distribution.
Let X be a discrete random variable. The
moment generating function MX ( t ) is
defined by For X and Y independent, MX +Y (t ) = MX (t ) MY (t ).
MX (t ) = E(e t X ) = e t x p(x)

169 172

Finding the nth moment: Exercise:

Let X1, X 2 , X 3 be a random sample from a
Finding the nth moment using discrete distribution with probability function
the moment generating function:
1
MX ( n ) (0) = x n p(x) = E(X n ) 3 for x = 0

2
p ( x) = for x = 1
3
0 otherwise

Determine the moment generating function,
170 M(t ) of Y = X1 X 2 X 3 . 173

Discrete Random Variable Example: Exercise, cont.:

(A) 19 + 8 e t
x 0 1 2 27 27
p(x) .5 .3 .2 (B) 1 + 2e t
etx 0t
e =1 e1t e2t 3
1 2 t
(C) + e
MX ( t ) = 1( .5) + e t ( .3 ) + e 2t ( .2 ) 3 3
1 8 3t
MX ( t ) = 0 + e t ( .3 ) + e 2t ( .2 )( 2 ) (D) + e
27 27
MX ( 0 ) = 0 + 1( .3 ) + 1( .2)( 2 ) = E ( X ) 1 2 3t
(E) + e
171 3 3 174

Copyright ACTEX 2006

 Solution: Solution, cont.:
Since each X i can be only 0 or 1, the product
Y = X1, X 2 , X 3 can be only 0 or 1. In addition, MX ( 0 ) = 10, 000
Y is 1 if and only3if all of the X i are 1. Thus MX ( 0 ) = 125, 000, 000
2 8
P(Y = 1) = =
3 27 V (X ) = E (X 2 ) E (X )
2
3
2 19
P(Y = 0) = 1 P(Y = 1) = 1 = = 125, 000, 000 10, 000 2
3 27
19 0t 8 1t 19 8 t = 25, 000, 000
MY ( t ) = E ( e Yt ) = e + e = + e
27 27 27 27
X = V ( X ) = 25, 000, 000 = 5, 000
Answer A 175 Answer B 178

Exercise: Exercise:
An actuary determines that the claim size for a A company insures homes in three cities, J,
certain class of accidents is a random variable, K, and L. Since sufficient distance separates
X, with moment generating function the cities, it is reasonable to assume that the
1 losses occurring in these cities are
MX ( t ) = independent.
(1 2500t )
4
The moment generating functions for the
Determine the standard deviation of the claim loss distributions of the cities are:
MJ ( t ) = (1 2t )
3
size for this class of accidents.
MK ( t ) = (1 2t )
2.5
(A)1,340 (B) 5,000 (C) 8,660
(D) 10,000 (E) 11,180 ML ( t ) = (1 2t )
176 4.5 179

Solution: Exercise, cont.:

Use the derivatives of the moment generating
Let X represent the combined losses from the
function to find the first two moments and thus
three cities. Calculate E ( X 3 ) .
obtain V ( X ) = E ( X 2 ) E ( X ) .
2

MX ( t ) = (1 2500t )
4

(A)1,320 (B) 2,082 (C) 5,760

MX ( t ) = 4 (1 2500t ) ( 2500 )
5
(D) 8,000 (E) 10,560

= 10, 000 (1 2500t )

5

MX ( t ) = 50, 000 (1 2500t )

( 2500 )
6

= 125, 000, 000 (1 2500t )

5
177 180

Copyright ACTEX 2006

 Solution: Exercise:
Recall that E ( X ) = MX ( 0 ) . First find MX ( t ) .
3 A car dealership sells 0, 1, or 2 luxury
Note that X = J + K + L where summands are cars on any day. When selling a car, the
independent. Thus dealer also tries to persuade the customer to
MX ( t ) = M J+ K +L ( t ) buy an extended warranty for the car.
Let X denote the number of luxury cars
= M J ( t ) M K ( t ) ML ( t ) sold in a given day, and let Y denote the
= (1 2t )
3
(1 2t )
2.5
(1 2t )
4.5 number of extended warranties sold.

= (1 2t )
10

181 184

Solution, cont.: Exercise, cont.:

MX ( t ) = 10 (1 2t )
11
( 2) = 20 (1 2t )
11 P(X = 0, Y = 0) = 1/ 6
P(X = 1, Y = 0) = 1/12
MX ( t ) = 220 (1 2t ) ( 2) = 440 (1 2t ) P(X = 1, Y = 1) = 1/ 6
12 12

P(X = 2, Y = 0) = 1/12
MX ( t ) = 12 ( 440 )(1 2t ) ( 2 )
13
P(X = 2, Y = 1) = 1/ 3
P(X = 2, Y = 2) = 1/ 6
= 10, 560 (1 2t )
13

What is the variance of X?

E ( X 3 ) = MX ( 0 ) = 10, 560
(A) 0.47 (B) 0.58 (C) 0.83 (D) 1.42 (E) 2.58
Answer E 182 185

Discrete Joint Probability Function: Solution:

y Definition:
Let X and Y be discrete random First put the given information into a bivariate
variables. The joint probability function table and fill in the marginal probabilities for X.
for X and Y is the function 0 1 2
p(x , y ) = P(X = x, Y = y ). X
Y
Note that: p(x, y) = 1 0 1/6 1/12 1/12
x y
1 0 1/6 1/3
y The
Definition:
marginal probability functions of X 2 0 0 1/6
and Y are defined by 1/6=2/12 3/12 7/12
p X ( x) = p ( x , y ) pY (y ) = p(x, y ) pX ( x)
183 186
y x

Copyright ACTEX 2006

 Solution, cont.: Definition:
The joint probability density function for two
2 3 7 17 continuous random variables X and Y is a
E ( X ) = 0 + 1 + 2 =
12 12 12 12 continuous, real valued function f(x,y) satisfying:
2 3 7 31 iii) P(a X b, c Y d) is given by the
E ( X 2 ) = 0 2 + 12 + 22 = volume between the surface z = f(x,y)
12 12 12 12
and the region in the x-y plane bounded
2
31 17 by x = a , x = b, y = c and y = d.
V (X ) = = .576
12 12 b d

Answer B P(a X b , c Y d) = f (x , y ) dy dx
a c
187 190

Definition: Definition:
The joint probability density function for two Let f(x,y) be the joint density function for the
continuous random variables X and Y is a continuous random variables X and Y. The
continuous, real valued function f(x,y) satisfying: marginal distribution functions of X and Y
i) f(x,y) 0 for all x,y. are defined by:

fX ( x ) = f(x, y)dy


fY ( y ) = f(x, y)dx

188 191

Definition:
The joint probability density function for two
continuous random variables X and Y is a
continuous, real valued function f(x,y) satisfying:
ii) The total volume bounded by the graph of
z = f(x,y) and the x-y plane is 1.
f (x, y) dx dy = 1
-
Exercise:
A device contains two components. The
device fails if either component fails. The joint
density function of the lifetimes of the
components, measured in hours, is f(s,t), where
0 < s <1 and 0 < t <1 .
What is the probability that the device
fails during the first half hour of operation?

189 192

Copyright ACTEX 2006

 Exercise, cont.: Exercise:
0.5 0.5 The future lifetimes (in months) of two
(A) f ( s, t ) ds dt components of a machine have the following
0 0
1 0.5 joint density function:
(B) f ( s, t ) ds dt
0 0
6 ( 50 x y )
1 1
f ( s, t ) ds dt , 0 < x < 50 y < 50
(C) 0.5 0.5 f ( x , y ) = 125, 000
0, otherwise
0.5 1 1 0.5

(D) f ( s, t ) ds dt +
0 0 0 0
f ( s, t ) ds dt
0.5 1 1 0.5
(E) f ( s, t ) ds dt +
0 0.5 0 0
f ( s, t ) ds dt What is the probability that both components
are still functioning 20 months from now?
193 196

Solution: Exercise, cont.:

The device fails if either S < 1/ 2 or T < 1/ 2. 6
( 50 x y ) dy dx
20 20

125, 000 0 0

(A)
S 6 30 50 x
( 50 x y ) dy dx
125, 000 20 20

1 (B)
A 6 30 50 x y
( 50 x y ) dy dx
125, 000 20 20

(C)
.5
6 50 50 x
( 50 x y ) dy dx
125, 000 20 20

B (D)

6 50 50 x y
.5 1 T ( 50 x y ) dy dx
125, 000 20 20
(E)
194 197

Solution, cont.: Solution:

Upper limits of integration in choices C and E
P ( S < 1/ 2 or Y < 1/ 2 ) are clearly incorrect.

= f ( s , t ) ds dt + f ( s , t ) ds dt We need P ( X 20 & Y 20 ) from A, B or D.

A B
0.5 1 1 0.5 Density function is non-zero only in the first
=
0 f ( s , t ) ds dt + f ( s , t ) ds dt
0.5 0 0 quadrant triangle bounded above by the line
x + y = 50 or y = 50 x.

Answer E
195 198

Copyright ACTEX 2006

 Solution, cont.: Solution:
In the diagram, below, we show the triangle The device fails if either X < 1 or Y < 1.
and the region R where both components are The set of pairs (x,y) for which this occurs is
still functioning after 20 months. shown in the shaded region in the diagram
y below. y
50 2
B
R 1
20 A

20 30 50 x 199
1 2 x 202

Solution, cont.: Solution, cont.:

y
P ( X 20 & Y 20 ) 2 For the shaded region A,
P ( X < 1 or Y < 1)
= f ( x , y ) dy dx B
= f ( x , y ) dx dy
R

6 30 50 x
1
A
20 ( 50 x y ) dy dx
A
=
125, 000 20

1 2 x
Answer B

200 203

Exercise: Solution, cont.:

A device runs until either of two components y
Integrate over the
fails, at which point the device stops running. The 2
unshaded rectangle B,
joint density function of the lifetimes of the two B to get the
components, both measured in hours, is 1 complementary
x+y A
f ( x, y ) = for 0 < x < 2 and 0 < y < 2 probability.
8
P ( X < 1 or Y < 1)
What is the probability that the device 1 2 x
fails during its first hour of operation? = 1 f ( x , y ) dx dy
B

(A) .125 (B) .141 (C) .391 (D) .625 (E) .875 2 2 x+y
= 1 dx dy
201 1 1
8 204

Copyright ACTEX 2006

 Solution, cont.: Exercise, cont.:
y 2x+y
2
Determine the probability that the portion
2
1 1
8
dx dy
of a claim representing damage to the house is
B 2
less than 0.2 .
1 2 x2
1 = + xy dy
A 8 1 2 1 (A) .360 (B) .480 (C) .488 (D) .512 (E) .520
1 2
= (1.5 + y ) dy
1 2 x 8 1
2
1 y2 3
= 1.5y + =
8 2 1 8
= .375 205 208

Solution, cont.: Solution:

y
Find P ( X < .2 ) = A f ( x, y ) dy dx where A is
2
the region indicated in the diagram below.
B y
1 1
A y = 1-x

1 2 x A

P ( X < 1 or Y < 1) = 1 .375 = .625

.2 1 x
Answer D 206 209

Exercise: Solution, cont.:

A company is reviewing tornado damage y
claims under a farm insurance policy. Let X be 1
y = 1-x
the portion of a claim representing damage to
the house and let Y be the portion of the
f ( x, y ) dy dx
A
1 x
= 6 [1 x y ] dy dx
.2
same claim representing damage to the rest of A
0 0
the property. The joint density function of X and
Y is
6 1- ( x + y ) , x > 0, y > 0, and x+y < 1 .2 1 x
f ( x, y ) =
0, otherwise
207 210

Copyright ACTEX 2006

 Solution, cont.: Definitions:
1 x
y y Two discrete random variables X and Y are
2
1 x
[1 x y ] dy dx = 6
.2 .2
6 y xy dx
0 0 0
2 0 independent if
p(x , y ) = p X (x) pY (y )
.2 (1 x ) dx
2

= 6 (1 x ) x (1 x ) for all pairs of outcomes (x,y).

0 2

.2 y independent
Two continuous random variables X and Y are
.2 (1 x ) (1 x )
2 3
if
= 6 dx = 6 = .488 f (x , y ) = fX (x) fY (y )
0 2 6 0
Answer C for all pairs (x,y).
Note: We can also work this using the marginal
211 214
for X. The calculations are basically the same.

Definitions: Exercise:

y Discrete Case. The conditional distribution of X A diagnostic test for the presence of a
disease has two possible outcomes: 1 for
given that Y=y is given by
p (x , y ) disease present and 0 for disease not
P(X = x |Y = y ) = p(x | y ) = . present. Let X denote the disease state of a
pY (y )
patient, and let Y denote the outcome of the
y Continuous Case. Let X and Y be continuous diagnostic test.
random variables with joint density function
f(x,y). The conditional density for X given
that Y=y is given by
f (x , y )
f (x |Y = y ) = f (x | y ) = .
fY (y ) 212 215

Conditional Expected Value: Exercise, cont.:

y For discrete random variables, The joint probability function of X and Y
is given by:
E(Y | X = x) = y p(y | x)
y P(X = 0, Y = 0) = 0.800
E(X |Y = y ) = x p(x | y ) P(X = 1, Y = 0) = 0.050
x P(X = 0, Y = 1) = 0.025
y When X and Y are continuous, the conditional P(X = 1, Y = 1) = 0.125
expected values are given by Calculate Var( Y | X =1) .

E(Y | X = x) = y f (y | x) dy (A) 0.13 (B) 0.15 (C) 0.20 (D) 0.51 (E) 0.71


E(X |Y = y ) = x f (x | y ) dx
213 216

Copyright ACTEX 2006

 Solution: Exercise:
We can calculate this variance if we know the Once a fire is reported to a fire insurance
conditional distribution of Y given that X=1. company, the company makes an initial
X 0 1 estimate, X, of the amount it will pay to the
Y claimant for the fire loss. When the claim is
0 .800 .050 finally settled, the company pays an amount, Y,
to the claimant. The company has
1 .025 .125
determined that X and Y have the joint density
pX ( x) .825 .175 function

( 2 x 1)
2
f ( x, y ) = 2 y ( ) , x > 1, y > 1
x 1

x ( x 1)
217 220

Solution, cont.: Exercise, cont.:

P (Y = 0 & X = 1)
P (Y = 0| X = 1) = Given that the initial claim estimated by
P ( X = 1) the company is 2, determine the probability
.05 that the final settlement amount is between 1
= = .2857 and 3.
.175
P (Y = 1& X = 1) A) 1/9 B) 2/9 C) 1/3 D) 2/3 E) 8/9
P (Y = 1| X = 1) =
P ( X = 1)
.125
= = .7143
.175
218 221

Solution, cont.: Solution:

To find P (1 < Y < 3| X = 2 ) we need:
Use V (X ) = E ( X 2 ) E ( X ) .
2
f ( 2, y ) .5y 3
f ( y |X = 2) = =
fX ( 2 ) fX ( 2 )
E (Y | X = 1) = .2857 ( 0 ) + .7143 (1) = .7143 y 2

1
fX ( 2 ) = f ( 2, y ) dy = .5y 3dy = =
1 1 4 4
E (Y 2 | X = 1) = .2857 ( 0 ) + .7143 (1) = .7143 .5y 3 .5y 3
2 2 1
f ( y | X = 2) = = = 2y 3

fX ( 2 ) (1/ 4 )
3
V ( X ) = .7143 ( .7143 ) = .204
2
P (1 < Y < 3| X = 2 ) = f ( y | X = 2 ) dy
1
Answer C 3 3 8
= 2y 3dy = y 2 =
Answer E 9
219 222
1 1

Copyright ACTEX 2006

 Review: Exercise, cont.:
What is the probability that these 4 will
Counting Partitions: contain at least two more high-risk drivers than
The number of partitions of n objects into k low-risk drivers?
distinct groups of size n1, n2 , , n k is given by
(A) .006 (B) .012 (C) .018 (D) .049 (E) .073
n n!
n , n , ..., n =
1 2 k n1 ! n2 !... n k !

223 226

Review, cont.: Solution:

The Multinomial Distribution: L, M, and H: number of low risk, moderate
Random experiment has k mutually exclusive risk and high risk drivers respectively.
outcomes E1, , E k , with P(E i ) = p i . Repeat p1 = P(L) = .50
this experiment in n independent trials. p 2 = P(M) = .30
Let X i be the number of times that the outcome p 3 = P(H) = .20
E i occurs in the n trials.
P ( X1 = n1 & X 2 = n2 & ... & X k = n k ) There are four cases:

n n1 n2
= p1 p 2 ... p k
nk

n ,
1 2 n , ..., n k 224 227

Exercise: Solution, cont.:

A large pool of adults earning their first drivers
1 P (L = 0 & M = 0 & H = 4) = 0, 0, 4 .5 .3 .2
4 0 0 4
license includes 50% low-risk drivers, 30%
moderate-risk drivers, and 20% high-risk
drivers. Because these drivers have no prior = 1( .50.30.2 4 )
driving record, an insurance company considers
= .0016
each driver to be randomly selected from the
4 0 1 3
pool. This month, the insurance company writes 4
new policies for adults earning their first drivers 2 P ( L = 0 & M = 1& H = 3 ) = .5 .3 .2
0,1, 3
license.
= 4 ( .50.31.23 )
225
= .0096 228

Copyright ACTEX 2006

 Solution, cont.: Covariance Properties:

3 4 1 0 3
P ( L = 1& M = 0 & H = 3 ) = .5 .3 .2
y Covariance of X and Y:
1, 0, 3 Cov(X , Y ) = E ( X X )(Y Y )

= 4 ( .51.30.23 )
= .0160 y Alternative Calculation:
Cov(X , Y ) = E(XY ) - E(X ) E(Y )
4 0 2 2
4 P ( L = 0 & M = 2 & H = 2) = .5 .3 .2
0, 2, 2
= 6 ( .50.32.22 )
y X and Y Independent:
Cov(X , Y ) = 0
= .0216 229 232

Solution, cont.: Variance Properties:

The sum of these probabilities is .0488. y Variance of X + Y:

V (X + Y ) = V (X ) + V (Y ) + 2Cov(X , Y )
Answer D
y Variance of X+Y when
X and Y Independent:
V (X + Y ) = V (X ) + V (Y )

230 233

Expected Value Properties: Useful Properties of Covariance:

y Sum of Two Random Variables:
1 Cov(X , Y ) = Cov(Y , X )
E(X + Y ) = E(X ) + E(Y )

y Product of Two Random Variables: 2 Cov(X , X ) = V (X )

Discrete Case:
E [(XY )] = (xy ) p(x, y ) 3 If k is a constant random variable, then
x y Cov(X , k) = 0.
Continuous Case:
E [(XY )] =

xy f ( x, y ) dy dx 4 Cov(aX , bY ) = ab Cov(X , Y )

y Product when X and Y Independent: 5 Cov(X , Y + Z) = Cov(X , Y ) + Cov(X , Z)

E(XY ) = E(X ) E(Y ) 231 234

Copyright ACTEX 2006

 Correlation Coefficient: Solution:
A) Look at the total hours for a single
Cov ( X , Y ) individual
X ,Y = , 1 X ,Y 1
X Y B) Use the central limit theorem and normal
approximation.
For one individual, the total hours watching
movies or sporting events is S = X + Y .
E ( S ) = E ( X + Y ) = E ( X ) + E (Y ) = 50 + 20 = 70

V ( S ) = V ( X + Y ) = V ( X ) + V (Y ) + 2Cov ( X , Y )
235
= 50 + 30 + 2 (10 ) = 100 238

Exercise: Solution, cont.:

Let X and Y be the number of hours that a One hundred people are assumed iid. The total
randomly selected person watches movies and for all 100 people is T = S1 + ... + S100 .
sporting events, respectively, during a three- By the central limit theorem, T is approximately
month period. The following information is known normal with
about X and Y: E(T ) = T = 100(70) = 7, 000
V (T ) = T 2 = 100(100) = 10, 000
E ( X ) = 50 Var ( X ) = 50
S = 10, 000 = 100
E (Y ) = 20 Var (Y ) = 30
7,100 7, 000
Cov ( X , Y ) = 10 Thus, P ( T < 7100 ) = P Z <
100
236 = P ( Z < 1) = .8413 239
Answer B

Exercise, cont.: Exercise:

One hundred people are randomly selected Let X and Y be continuous random variables
and observed for these three months. Let T be with joint density function
the total number of hours that these one hundred 8
people watch movies or sporting events during xy for 0 x 1, x y 2x
f ( x, y ) = 3
this three-month period. 0 otherwise
Approximate the value of P(T < 7100).
Calculate the covariance of X and Y.
(A) 0.62 (B) 0.84 (C) 0.87 (D) 0.92 (E) 0.97
(A) 0.04 (B) 0.25 (C) 0.67 (D) 0.80 (E) 1.24
237 240

Copyright ACTEX 2006

 Solution: Solution, cont.:
8 1 2x 2
Cov ( X , Y ) = E ( XY ) E ( X ) E (Y ) E ( X ) = x f ( x , y ) dy dx =
3 0 x
x y dy dx
R

8 1 2 y 2
2x
y y = 2x
3 0 2 x
= x dx
y=x
2
8 3
1
= x 4 dx
1 A 3 0 2
1
x5 4
x = 4 =
1 241
5 0 5 244

Solution, cont.: Solution, cont.:

8 1 2x 2 2 Cov ( X , Y ) = E ( XY ) E ( X ) E (Y )
E ( XY ) = xyf ( x , y )dydx =
3 0 x
x y dy dx
R
56 4 56
=
8 y3
2x
1 8 1 7 54 5 45
= x2 dx = x 5 dx
3 0 3 3 0 3 = .041
x

1
Answer A
56 x 6 56
= =
9 6 0 54

242 245

Solution, cont.: Exercise:

8 1 2x 2
E (Y ) = y f ( x , y ) dy dx =
3 0 x
xy dy dx X: Size of a surgical claim
R Y: Size of the associated hospital claim
8 1 y 3
2x
8 1 7
= x dx = x 4 dx An actuary is using a model in which
3 0 3 3 0 3
x E(X) = 5,
1 E(X2) = 27.4,
56 x 5 56 E(Y) = 7,
= =
9 5 0 45 E(Y2) = 51.4, and
V(X+Y) = 8.
243 246

Copyright ACTEX 2006

 Exercise, cont.: Theorem:
Let C1 = X+Y denote the size of the combined
claims before the application of a 20% Double Expectation
surcharge on the hospital portion of the claim, Theorem of the Mean:
and let C2 denote the size of the combined E[E(X |Y )] = E(X ) and E[ E(Y | X )] = E(Y )
claims after the application of that surcharge.
Calculate Cov(C1, C2).

(A) 8.80 (B) 9.60 (C) 9.76 (D) 11.52(E) 12.32

247 250

Solution:
C2 = X + 1.2Y
Cov(C1, C2 ) = Cov ( X + Y , X + 1.2Y )
= Cov(X , X ) + Cov(X ,1.2Y ) + Cov(Y , X ) + Cov(Y ,1.2Y )
= V (X ) + 1.2Cov(X , Y ) + Cov(X , Y ) + 1.2Cov(Y , Y )
= V (X ) + 2.2Cov(X , Y ) + 1.2V (Y )
V ( X ) = E ( X 2 ) E(X )2 = 27.4 52 = 2.4
V (Y ) = E (Y 2 ) E(Y )2 = 51.4 7 2 = 2.4
V ( X + Y ) = V ( X ) + V (Y ) + 2Cov ( X , Y ) 248
Exercise:
An auto insurance company insures an
automobile worth 15,000 for one year under a
policy with a 1,000 deductible. During the
policy year there is a 0.04 chance of partial
damage to the car and a 0.02 chance of a total
loss of the car.
251

Solution, cont.: Exercise, cont.:

8 = 2.4 + 2.4 + 2Cov ( X , Y ) If there is partial damage to the car, the

amount X of damage (in thousands) follows a
Cov ( X , Y ) = 1.6 distribution with density function

Now we have the required information. 0.5003e x / 2 for 0 < x < 15

f ( x) =
0 otherwise
Cov(C1, C2 ) = V (X ) + 2.2Cov(X , Y ) + 1.2V (Y )
= 2.4 + 2.2 (1.6 ) + 1.2 ( 2.4 ) = 8.80 What is the expected claim payment?

(A) 320 (B) 328 (C) 352 (D) 380 (E) 540
Answer A
249 252

Copyright ACTEX 2006

 Solution:
There are three possible cases. Amounts are
expressed in thousands.

a) No damage.
P ( No Damage ) = 1 .04 .02 = .94
E ( Amount Paid|No Damage ) = 0
b) Full damage.
P ( Total Loss ) = .02
E ( Amount Paid|Total Loss ) = 15 1 = 14 253

Solution, cont.:
c) Partial damage.
P ( Partial Damage ) = .04
E ( Amount Paid|Partial Damage )

( x 1) f (x) dx
15
=
1

( x 1) e x / 2dx = 1.2049
15
= .5003
1
Expected amount paid in thousands
.94 ( 0 ) + .02 (14 ) + .04 (1.2049 ) = .328
Answer B 254

Copyright ACTEX 2006