FINA2220A

Quantitative Methods for

Actuarial Analysis I

Chapter 6
Jointly Distributed Random Variables

⼤階 F vs
細階 f ?
" (x ≤
)
functio .
 Joint Distribution Functions
 There are probability statements concerning two or more
random variables
 In order to deal with such probability, we define the joint
cumulative probability distribution function
 For any two random variables X and Y, the joint cumulative
probability distribution function (joint c.d.f.) is defined as

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 2

 Joint Distribution Functions
 The distribution of X can be obtained from the joint distribution
of X and Y as follows:

 Similarly

 The distribution FX and FY are referred to as the marginal

distributions of X and Y
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 3
 Joint Distribution Functions
 All joint probability statements about X and Y can be answered
in terms of their joint distribution function
 Suppose we want to compute the joint probability that X is
greater than a and Y is greater than b, we have
√

)
√

…
P( x ε a
,
Y≤ b
A more general equation for a1< a2, b1< b2 is

Complicated
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 4
P ( XEan Y ≤
,
| tp ( XEa
kabb → P,
( X ≤ b . ) Plx ≤ ib -
0
, , a ,
≤
 Joint Distribution Functions
 In the case when X and Y are both discrete random variables, it
is convenient to define the joint probability mass function of X and
Y by 細皆 pp

 The probability mass function of X can be obtained from p(x, y)

by PCANB )
=
X = x ↑= O
, 1,2
…
∞
,

 Similarly

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 5

 Exercise 6.1
 At Simple Choice Insurance Company, customers choose from 3
levels of automobile liability coverage: 1, 3 or 5 million and 3
levels of personal liability coverage: 0, 1 or 4 million
 Let X be the level of auto liability selected (in millions)
 Let Y be the level of personal liability selected (in millions)
 The joint probability mass function is shown in the following table
y
p(x,y) )
0 1 4 P( x

1 0.10+
+
0.05 ⼗
0.15⼗
=
O , 30
1
+

x 3 0.05 0.20 0.25

⾶ )
t f C 0. 5 5 4 =
f f ⼗ ,

5 0.15 ⼠⼗ 0.00 0.05 =

0 20
PCy) 45
.

0 O 25 0
.

30 .
,.

1 00
 Verify that this is in fact a joint probability mass function
.

⼯
 Calculate P(X = 3, Y = 4) 0 25≈
-
Sum ofpk y) ,

 Write the marginal distributions for X and Y = 1

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 6

 Out of syllabus SOA
Joint Probability Density Functions
X and Y are jointly continuous if there exists a function f(x, y)
defined for all real x and y, having the property that for every set
C of pairs of real numbers
c
”

 The function f(x, y) is called the joint probability density function

(joint p.d.f.) of X and Y
 If A and B are any sets of real numbers, then we have

b
B =
( b , ,
.
)2 -

b ,
⼀

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables

xdd Page 7

A =
( a
,
az
)
 Joint Probability Density Functions
 The joint c.d.f. can be written as P(
x ≤
njy ≤ b )

 By differentiation, we have Constant

tkanltady
I

最哥 al . "
. '

wherever the partial derivatives exists —

了。 fcy ) dy
 The joint density function can also be defined as follows:
g( b )
=

{f ( x, bldx
when da and db are small and f(x, y) is continuous at a, b f =
)ba
,

 Hence, f(a, b) is a measure of how likely it is that the random

vector (X, Y) will be near (a, b)

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 8

 Joint Probability Density Functions
 If X and Y are jointly continuous, they are individually
continuous, and their probability density functions can be
obtained in a similar way as that in the discrete case:

 Thus the (marginal) p.d.f. of X is given by

 Similarly, the p.d.f. of Y is given by

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 9

 Exercise 6.2
 Suppose that random variables X and Y are jointly distributed
on the rectangle R given by [0, 2]  [0, 5], with joint density
function

 Find the marginal probability density functions of X and Y

 Find P(X + 2Y > 2)
fx 4) = 5
-f (x ,
4
dy =
)
[

501 dy= o -
ly 18 = 0 5 .
0 ≤ x ≤2

f0
=S
fyl ) f (xy| dx . 1
=
,
δ 2
02
dx
011 O≤
=

Y ε<
. 0

Y
X and are marg
'inad unformly distribuled

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 10

 surfacerplane e
C .
15

「 →
C (O,

f (x
,

)1 )
。 C .
05

@ O .
O0
Z

s
/ /
/ / i
×
4
23 /
0
Me thod 1 ( for thil speclal case only ) ;

Let A be the regm bounded by

oc ≤ 2
yx
0
Ey ≤ and xt ,
2
5

Y
5) x 2
5 AreaotregionA = +
”

ximiil Plxlene 92 )

l' pidf surfoo

unlher the
the
ovea reg7 on A
,
= O . ( x Area of regin A

N
( = C , 1 x 9
×
0 i …
=
6 9 ,

.
Method 2 : double integration in onder of x andy
X > 2
-

2 个

P (x + 2 Y > 2 )
Y Keep
1 0 fttldady
f
: constant
5 。

"
2 2

i
-

filiiyoidadytf s oidady
= 5

(
tosg ,
…
=
fi '
locx zzdy tf
]
,

5
'

[ ox
]
.

dy
bMtl
x +2
'

y 2
So ozydy tf 50 24 dy
=
1 = .

∴←imycan
,

olyli ylir
µ
t o 2

" ×
0 z ….
= O .
1 + 0 . 8
.
=
0 9 .
Method 3 : double integration in order of yand x

y P (X +2 Y >2 ) =
f aJi flx yldy↓ ,
.

- xi 2

5 :

fi 'lixs oildyta
Regim ! P
filoylits =
dx

{ 01 ( 49 ± ) ↓=
1
。

×+￥ )
f( x
) =
0 - (4
y
,

— ,

|
6 =
O 1 (
1

)
=

: 8+1
.
,

∞
=
0 ,
9
1 1 -

xV 25
y
( <

:
×
0 ←
之
fixed betveen oand 2
 Method 2
, eduivalent
3

In Method 3 ,

O ≤ ≤ 2
OEY ≤
X

{ ⼀会
5

0
「 EYE 5 { 2 2y-
≤ x ≤2
cx ≤ 2 1 ≤
=> y ≤5

{
,

2
2 -
2 ≤≤
x 2 O ≤
f
≤1 xwhenx
2
<≤
=
0

y ,

as f( x ) =
O for x < z
, y r <
y =
- _

y
 Exercise 6.3
 Suppose that random variables X and Y are jointly distributed
on the rectangle R given by [0, 2]  [0, 5], with joint density
function

 Find the marginal probability density functions of X and Y

 Find P(X + 2Y > 2) O ≤ X ≤ 2
xy 15
fofexxdy 5 xhndy 0 . 0 5
fxlx 1
= =

=
= | . x

f
.

ficyS 0f ) = . ? xl -
0 ≤
y ≤5
onhoddx
'
f 005
,

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 15

 6 .

^ 0 .
2
.

1
0 .

0 .

coltxylixn
ly
p( x +
2
) f ' fiin oilidyttf . ↓
>
=
.
'

= S .
0 .
1 (4
τ ) ↓↓= 1 ☆^+÷)↑=±=
xt 0
. (
。
0 .
933337
 Joint Distribution of Several Random Variables
 We can also define joint probability distributions for n random
variables
 The joint cumulative probability distribution function
F(a1, a2, …, an) of n random variables X1, X2, …, Xn is defined by

 Further, the n random variables are said to be jointly continuous

if there exists a function f(x1, x2, …, xn), called the joint
probability density function, such that for any set C in n-space

 In particular, for any n sets of real numbers A1, A2, …, An

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 17

 The Multinomial Distribution
 One of the most important joint distributions is the multinomial
 This arises when a sequence of n independent and identical
experiments is performed S M L X 1
,
X×x 2
, , ,
x×L
,

 Suppose that each experiment can result in any one of r possible

6
outcomes, with respective probabilities
r =

=| P P P P
-

P P
- - -
-

2
1 , 4 5
6

 If we let Xi denote the number of the n experiments that result in

outcome number i, then

pla succes, n -

n
failm )
 Here,
,

ifr
,

: 2
nn

(∴ ,
,

)
,
=
) P
" ( 1

- p

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 18

 Independent Random Variables
 Two random variables X and Y are said to be independent if for
any two sets of real numbers A and B For event E ond F ,

P =Λ F )
(5 =
PCE) PC )

 This is the condition of independence

 It can be shown by using the three axioms of probability that the
above equation will follow if and only if for all a, b

 Hence, in terms of the joint distribution function F of X and Y,

we have that X and Y are independent if

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 19

 Independent Random Variables
 When X and Y are discrete random variables, the condition of
independence is equivalent to

 The equivalence can be shown as follows

 If we let A and B in the condition of independence be the one point
sets A = {x} and B = {y}, then the above equation follows
 If the above equation is valid, then for any sets A, B
l
Ifunction
oty) independentx
Coust intan
inner sumomation
e

inner Summatiou

Ʃ Pyly ) PCXAt
…

constant

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 20

 Independent Random Variables
 In the jointly continuous case, the condition of independence is
equivalent to

 Loosely speaking, X and Y are independent if knowing the value

of one does not change the distribution of the other
 Random variables that are not independent are said to be
dependent

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 21

 Example 6.1
 Suppose that n + m independent trial, having a common success
probability p, are performed
 Let X be the number of successes in the first n trials
 Let Y be the number of successes in the final m trials
 Then it is easy to show that X and Y are independent

 LetZ be the total number of successes in the n + m trials

 Then it can be shown that X and Z are dependent
2 = ntm
, λ= n

X
1
2=
-
= n
or mta - t
,

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 22

 Exercise 6.4
 An insurance company sells two types of auto insurance policies:
Y X

Basic and Deluxe

<

X !
1
1

 The time until next Basic Policy claim is an exponential random

variable with mean two days Ψ
 The time until the next Deluxe Policy claim is an independent
exponential random variable with mean three days
 What is the probability that the next claim will be a Deluxe
Policy claim? Px( )
5
±
fopeluxe
Basc

te je
x "
( ) ( y)
-

f and
x =

0 Y 30
=

< xs
, ,

P ( Next claim is a Deluxe Polcy cdam ) =

P( X 3 )

,
1 ,fanidady : 10 gtx fpcy, dxdy= .fy0te ' '" fxdy
'
jle
f "

= f t

.
2
]
ydy = ( - . ☆= 5
"

, Ie
^{ "
↑ y ∞je - -
± t

=
0 j

1
=

{
e

] - +
δ
( e i
= 。 。

。
-
e

" "

, e
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 23
 Independent Random Variables
 Proposition 6.1:
 The continuous (discrete) random variable X and Y are independent
if and only if their joint probability density (mass) function can be
expressed as

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 24

 Exercise 6.5
 If the joint density function of X and Y is

P =
2 P( 2 )

20 1
 Are the random variables independent?
=

2
P( 1 )
. 1 )

) 4x e 2 e y fx ( ) fyly )
2 =
2 2 x
.

=
f(
- -

x
,
y .
= x =
2

* 1

(x )
-

fx 4×
'
2x 3
(

Le 2x
2

) X
)

Gammal )
=
e = 0 < xc ∞ = ~
3 2
,
^ ( 3 )

fyl) : 2 ery O <

y
CD

≥ 个~ Expcy

X anl Y are
independent

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 25

 Exercise 6.6
 If the joint density of X and Y is
上
⼀

Z &
x+ 1 1
-

三
⼀

==
 Are the random variables independent? 0 !

f ( xi ) cannot be factorized into 2 funclions

I (x ifO < xc 1 ,
ocyclyOcxtyc
Let =
'
1
,) {

0 otherwise

f ( xy) = factorinto part ,

24
xyI ( x ,
n )
does nol 2

X and Y ave not independent

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 26

Independence of Several Random Variables
 The concept of independence can be defined for more than two
random variables
 In general, the n random variables X1, X2, …, Xn are said to be
independent if, for all sets of real numbers A1, A2, …, An

 This condition is equivalent to

 Finally,we say that an infinite collection of random variables is

independent if every finite collection of them is independent

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 27

 Sums of Independent Random Variables
 In many situations, we need to know the distribution of X + Y
from the distribution of X and Y
 Suppose that X and Y are independent, continuous random
variables having p.d.f. fX and fY
 The c.d.f. of X + Y is obtained as follows:

~
Region C

constant

 The cumulative distribution function FX+Y is called the

convolution of the distribution FX and FY
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 28
 Sums of Independent Random Variables
 Differentiating the cumulative distribution function FX+Y, we
obtain the probability density function fX+Y of X + Y

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 29

Exercise 6.7 (Sum of Independent Uniform)
 Suppose that X1, X2, …, Xn are independent uniform (0, 1)
random variables
 Show that

Proof Tn induction
xEl l
For n =1 , Fcx )
=x ,O ≤ stoue

f (x ,
x. ) =|= ft ,
x2 ) ⼀入 *
,
: X
2

For i 1 lustration for O ≤x ≤ 1

,
, [ 以
FIx ) :
PCX
)

, tX 2 Et .
T [
* *

dx .
dx ,
! !,
! 1
.
*
=
9
。
( xx
.
) dx ,

《} 这
)
Ʃ
.
*

=
: =

。 ⼯ !

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 30

 Assme n true for
the Efuation e s
n
1
.

Wvite =Ʃ X
蛋 X: Fn ( x
) =

xitX
,

.
Cnr ) !

O ≤x
⾊
For ≤l
,
"
xittrt . ttny C

Y
~
%
Fmilay
fxitidy ) ! ∴ ! !
-

前: 1 " (x jn" dy
=

nC - cxy
。
]
± x
!
by induction
=
n

As the toue for

eqoatinon a n
, ,

The
equalun s true for anyaleger n
 Sum of Independent Gamma
 Proposition 6.2:
 If X and Y are independent gamma random variables with
respective parameters (s, ) and (t, ) common

 Then X + Y is a gamma random variable with parameters (s + t, )

 The proof can be based on the convolution
 Corollary 6.1:
 If Xi, i = 1, 2, …, n are independent gamma random variables with
respective parameters (ti , ), i = 1, 2, …, n
 Then the sum Y = X1 + X2 + … + Xn is gamma with parameters

 The density function is

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 32

 Chi-Squared Distribution
 Proposition 6.3: Z
;
~
Nco," ;
'
^
X
?= Cauma E , z )
 If Z1, Z2, …, Zn are independent standard normal random variables
 Then the sum is said to have the chi-squared, 2,
distribution with n degrees of freedom Gamma ( , ) i

 When n = 1, with probability density function

 This is just a gamma distribution with parameters (½ , ½)

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 33

 Chi-Squared Distribution
 As each is gamma (½ , ½), we obtain from Corollary 6.1 that
the 2 distribution with n degrees of freedom is just the gamma
distribution with parameter (n/2, ½)
 The p.d.f. is given by

 The chi-squared distribution often arises in practice as being the

distribution of the square of the error involved when one
attempts to hit a target in n-dimensional space when the
coordinate errors are taken to be independent unit normal
random variables
 It is also important in statistical analysis

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 34

 Exercise 6.8
 For independent standard normal random variables Z1, Z2, Z3
and Z4
 Find the following probabilities


FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 35

 Sum of Independent Normal
 Proposition 6.4:
 If Xi, i = 1, …, n are independent random variables that are
normally distributed with respective parameters

 Then the sum X1 + X2 +… + Xn is normally distributed with

parameters

 The proof can be based on the convolution and induction

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 36

 Exercise 6.9
A company manufactures a brand of light bulb with a lifetime in
months that is normally distributed with mean 3 and variance 1
 A consumer buys a number of these bulbs with the intention of
replacing them successively as they burn out
 The light bulbs have independent lifetimes
 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?

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 37

 Lognormal Distribution
 The random variable Y is said to be a lognormal random variable
with parameters  and  if loge(Y) or ln(Y) is a normal random
variable with mean  and variance 2
 That is, Y is lognormal if it can be expressed as

where X is a normal random variable

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 39

 Exercise 6.10
 The claim severity random variable for an insurance company is
lognormal, and the normally distributed exponent has mean 6.8
and standard deviation 0.6
 What is the probability that a claim is greater than 1750?

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 40

 Exercise 6.11
 Starting at some fixed time, let S(n) denote the price of a certain
security at the end of n additional weeks, n  1
 A popular model for the evolution of these prices assumes that
the price ratios S(n)/S(n – 1), n  1, are independent and
identically distributed lognormal random variables
 Assuming this model, with parameters  = 0.0165 and  = 0.0730
 What is the probability that
(a) The price of the security increase over each of the next two weeks?
(b) The price at the end of 5 weeks is higher than it is today?

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 41

 Sum of Independent Poisson
 IfX and Y are independent Poisson random variables with
respective parameters λ1 and λ2
 Then X + Y has a Poisson distribution with parameter λ1 + λ2

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 44

 Sum of Independent Binomial
 Suppose that X and Y are independent Binomial random
variables with respective parameters (n , p) and (m , p)
 Then the sum X + Y has a Binomial distribution with parameters
(n + m, p)

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 45

 Conditional Distributions: Discrete Case
 The conditional probability of E given F is defined, provided that
P(F) > 0, as

 If X and Y are discrete random variables, it is natural to define

the conditional probability mass function of X given that Y = y by

for all y such that pY(y) > 0

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 46

 Conditional Distributions: Discrete Case
 Similarly, the conditional probability distribution function of X
given that Y = y is defined, for all y such that pY(y) > 0, by

 If X is independent of Y, the conditional mass function and

distribution function are the same as the unconditional one
 Since

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 47

 Exercise 6.12 (Exercise 6.1 Continued)
 Given that X and Y are discrete random variables with joint
probability mass function:
y
p(x,y)
0 1 4
1 0.10 0.05 0.15
x 3 0.05 0.20 0.25
5 0.15 0.00 0.05

 Calculate the conditional distribution of X given Y = 4

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 48

 Exercise 6.13
 IfX and Y are independent Poisson random variables with
respective parameters λ1 and λ2
 Compute the conditional distribution of X, given that X + Y = n

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 49

Conditional Distributions: Continuous Case
 If X and Y have a joint probability density function f(x, y), then
the conditional probability density function of X, given that Y = y,
is defined for all values of y such that fY(y) > 0, by

 To motive this definition, multiply the left-hand side by dx and

the right-hand side by (dx dy)/dy

 In other words, for small values of dx and dy, fX|Y(x|y)dx

represents the conditional probability that X is between x and
x + dx, given Y is between y and y + dy
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 50
Conditional Distributions: Continuous Case
 If X and Y are jointly continuous, for any set A

 By letting A = (, a], we can define the conditional cumulative

distribution function of X, given that Y = y, is defined as

 Note that even though the event on which we are conditioning,

namely the event {Y = y}, has probability 0, we have been able to
give workable expressions for conditional probabilities
 If X is independent of Y, the conditional density of X given that
Y = y is just the unconditional density of X
 Since

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 51

 Exercise 6.14
 Let f(x,y) = kx be a joint density function on the region R in the
plane described by 0  x  y  1
 Calculate the value of the constant k
 Find the marginal density functions for X and Y
 Calculate the conditional density of Y given X = ¼
 Calculate
 Calculate
 Calculate the conditional density of X given Y = ¾

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 52

 The Bivariate Normal Distribution
 The random variables X and Y have a bivariate normal
distribution if, for constants x, y, x > 0, y > 0, 1 <  < 1, their
joint density function is given, for all  < x, y < , by

 The marginal distribution of X is normal with mean x and

variance
 Similarly, the marginal distribution of Y is normal with mean y
and variance

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 56

 The Bivariate Normal Distribution
 The bivariate normal distribution could be written with matrix
notation

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 57

 The Bivariate Normal Distribution
 Given that Y = y, the random variable X is normally distributed
with

 The proof is omitted

 It follows that the necessary and sufficient condition for the
bivariate normal random variables X and Y to be independent is
that  = 0
 Note that when  = 0, the joint density factors into two terms, one
depending only on x and the other only on y

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 58

 Exercise 6.15
 Suppose that in a certain population of college students, the
respective grade point averages (GPAs), says X and Y, in high
school and the first year in college have an approximate bivariate
normal distribution with parameters x = 2.9, y = 2.4 , x = 0.4,
y = 0.5, and  = 0.8
 Find the (unconditional) probability that the first year GPA in
college of a student lies between 2.1 and 3.3
 Suppose that a student has a high school GPA of 3.2, what is the
probability that his first year GPA in college lies between 2.1 and
3.3

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 59

 Order Statistics
 Let X1, X2, …, Xn be n independent and identically distributed,
continuous random variables having a common density f and
distribution function F
 Define

 The ordered values are known as the order

statistics corresponding to the variables X1, X2, …, Xn
FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 61
 Density Function of j-th Order Statistic
 The density function of the jth-order statistic X(j) can be obtained
by direct reasoning as follows
 In order for X(j) to equal x, it is necessary for
 j – 1 of the n values X1, …, Xn to be less than x
 n – j of the n values X1, …, Xn to be greater than x
 1 of them to be equal x
 Now the probability density that any given set of j – 1 of the Xi’s
are less than x, another given set of n – j are all greater than x,
and the remaining value is equal to x, equals

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 62

 Density Function of j-th Order Statistic
 There are

different partitions of the n random variables X1, …, Xn into the

three groups
 Hence the density function of X(j) is given by

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 63

 C.D.F. of j-th Order Statistic
 The cumulative distribution of X(j) can be obtained by integrating
its p.d.f.

 Alternatively, note that the j-th order statistic is less than or

equal to y if and only if there are j or more of the Xi’s that are
less than or equal to y
 As the number of the Xi’s that are less than or equal to y is a
binomial random variable with parameters {n, p = F(y)}, we have

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 64

 Exercise 6.16
 Claim amount for wind damage to insured homes are
independent random variables with common density function

 Here, x is the amount of a claim in thousands

 Suppose 3 claims will be made
 What is the expected value of the largest of the three claims?
 What is the expected value of the smallest of the three claims?

FINA2220A (2022-2023) Chapter 6 Jointly Distributed Random Variables Page 65