Professional Documents
Culture Documents
11 12 26 14 52 48 1291 Maris PDF
11 12 26 14 52 48 1291 Maris PDF
11 12 26 14 52 48 1291 Maris PDF
CONTENTS
UNIT II
TWO DIMENSIONAL RANDOM VARIABLES
CHAPTER I
Definition of Two dimensional random variables
Joint Distributions
Distribution function
Probability mass function
Probability density function
Marginal and Conditional Distributions
CHAPTER II
Covariance
Correlation
Regression
CHAPTER III
Transformation of Random Variables
TWO DIMENSIONAL RANDOM VARIABLES 2
UNIT III
TWO DIMENSIONAL RANDOM VARIABLES
CHAPTER I
Note
(i)If the possible values of (X , Y) are finite or countable infinite, (X ,Y) is called a two
dimensional discrete random variable.
(ii)If (X , Y) can assume all values in a specified region R in the xy-plane, (X ,Y) is called a
two dimensional continuous random variable.
TWO DIMENSIONAL RANDOM VARIABLES 3
(ii) p
i j
ij =1
(ii ) f ( x , y ) d x d y = 1 .
R
Note
d b
P{a X b, c Y d } = f ( x , y ) dx dy
c a
F ( x, y ) = pij
j i
Properties of F (x , y)
(i) F(- , y) = 0 = F( x , - ) and F ( , ) = 1
(ii) P {a < X < b , Y y} = F (b , y) F (a , y)
(iii) P{X x , c <Y <d} = F (x , d) F (x , c)
(iv) P{a < X < b , c < Y <d } = F (b, d) F(a, d) F(b , c) + F(a , c)
(v) At points of continuity of f (x , y)
2 F
= f ( x, y )
x y
Pi* = P ( X = xi ) = pij
j
is called the marginal probability function of X. It is defined for X=x1 , x2 , and denoted as
Pi*. The collection of pairs {Xi , Pi* }, i = 1, 2, 3, is called the marginal probability
distribution of X.
Similarly the collection of pairs {Yj , P*j }, j = 1,2,3, is called the marginal
Similarly, f Y ( y ) =
f ( x, y)dx is called the marginal density of Y
Note
P (a X b ) = P (a X b, < Y < )
b
= f ( x, y)dxdy
a
b
= f ( x, y )dy dx
a
b
= f x ( x)dx
a
Similarly,
d
P(c Y d ) = fY ( y)dy
c
Problems
1 The joint probability mass function (PMF) of X and Y is
P (x, y) Y 0 1 2
X 0 0.1 0.04 0.02
1 0.08 0.2 0.06
2 0.06 0.14 0.3
TWO DIMENSIONAL RANDOM VARIABLES 6
independent.
Answer
Y
P ( x, y ) 0 1 2 Pi* = P ( X = xi )
X 0 0.1 0.04 0.02 P ( X = 0 ) = 0.16
Now P [ X 1, Y 1] = P [ X = 0, Y = 0 ] + P [ X = 0, Y = 1] + P [ X = 1; Y = 0 ] + P [ X = 1; Y = 1]
Y
X 1 2 3
0 3K 6K 9K
1 5K 8K 11K
2 7K 10K 13K
Answer
P (x,y) be the probability mass function, we have
3 2
j =1
i=0
P ( xi , yi ) = 1
3 The following table represents the joint probability distribution of the discrete
random variable ( X , Y )
Y
1 2 3
X
1 1/12 1/6 0
2 0 1/9 1/5
3 1/18 1/4 2/15
(i) Evaluate the marginal distribution of X and Y.
(ii) Find the conditional distribution of X given Y = 2.
(iii) Find the conditional distribution of Y given X = 3.
(iv) Find P(X 2 , Y = 3).
(v) Find P(Y 2) , P(X + Y < 4).
Answer
X
1 2 3 p(y)
Y
1 1/12 1/6 0 1/4
2 0 1/9 1/5 14/45
3 1/18 1/4 2/15 79/180
p(x) 5/36 19/36 1/3 1
x 1 2 3
p(x) 5/36 19/36 1/3
Marginal distribution of Y is
TWO DIMENSIONAL RANDOM VARIABLES 9
y 1 2 3
p(y) 1/4 14/45 79/180
P(X = 1 Y = 2)
P(X = 1/ Y = 2) =
P(Y = 2)
0
=
14 / 45
=0
P(X = 2 Y = 2)
P(X = 2 / Y = 2) =
P(Y = 2)
1/ 9
=
14 / 45
= 5/14
P(X = 3 Y = 2)
P(X = 3 / Y = 2) =
P(Y = 2)
by
1/ 5
=
14 / 45
= 9/14
P(Y = y X = 3)
P(Y = y / X = 3) = , where y = 1 , 2 , 3
P(X = 3)
P(Y = 1 X = 3)
P(Y = 1/ X = 3) =
P(X = 3)
0
=
1/ 3
=0
P(Y = 2 X = 3)
P(Y = 2 / X = 3) =
P(X = 3)
1/ 5
=
1/ 3
= 3/5
P(Y = 3 X = 3)
P(Y = 3 / X = 3) =
P(X = 3)
2 / 15
=
1/ 3
= 2/5
TWO DIMENSIONAL RANDOM VARIABLES 11
1 1
= +
18 4
11
=
36
1 14
= +
4 45
101
=
80
P(X + Y < 4) is given by
P(X + Y < 4) = P(X = 1, Y = 1) + P(X = 1, Y = 2) + P(X = 2, Y = 1)
1 1
= +0+
12 6
1
=
4
Answer
y f ( x, y )
f = XY , where f X ( x ) is the marginal density function of X
x fX ( x)
x
1
f X ( x) =
f ( x, y ) dy = x 8 x ( x y ) dy
x
1 2 y2
= x yx
8 2 x
1 3 x3 3 x3
= x + x
8 2 2
x3
= ,0 < x < 2
8
1
x( x y)
y 8
fY / X = ,0 < x < 2 and x < y < x
x 1 3
x
8
= 0, otherwise
x y
y 2 , x < y < x
i.e., f = x
x 0
,otherwise
1 1 x
P(X+Y<1) = f ( x, y ) dydx
0 0
1 1 x
= 4 xydydx
0 0
1 x
y2
1
= 4 x dx
0 2 0
1 1
= 2 x (1 x ) dx = 2 x (1 2 x + x 2 ) dx
2
0 0
1
= 2 ( x 2 x 2 + x3 ) dx
0
1
x 2 2 x3 x 4
= 2 +
2 3 4 0
1 2 1 6 8 + 3 1
= 2 + = 2 =
2 3 4 12 6
e
(
x2 + y 2 ) ;x>0,y>0
f(x,y) = Kxy .
Find the value of K and prove also that X and Y are independent.
Solution :
Here the range space is the entire first quadrant of the xy plane
By the property of the joint. p. d.f., we have,
(
x2 + y 2 )
x >0 y > 0
Kxy e dxdy = 1
TWO DIMENSIONAL RANDOM VARIABLES 14
K ye y .xe x dxdy = 1
2 2
0 0
Put
x2 = t 2xdx = dt
K
y 2
Then ye e t dt.dy = 1
2 0 0
t
K y2 e
2 0
ye ( )0 dy = 1
1
K
ye (1)dy = 1
2
y
2 0
K
2 0
y2
ye dy = 1
Put
y2 = v 2ydy = dv
K 1 v
2 2 0
e dv = 1
K 1 e v
=1
2 2 1 0
K1
(1) = 1
2 2
K
=1 K = 4
4
TWO DIMENSIONAL RANDOM VARIABLES 15
(
x2 + y 2 )
= 4 x ye dy
0
( )
y2
ye
x2
= 4 xe dy
0
1
x2
= 4 xe et dt [ Q Put y 2 = t 2 ydy = dt ]
0 2
x2 e t
= 2xe
1 0
2
= 2xe x , x > 0
TWO DIMENSIONAL RANDOM VARIABLES 16
xe
y2 x2
= 4 ye dx
0
1
y2
= 4 ye et dt [ Q Put x 2 = t 2 xdx = dt ]
0
2
2 e t
= 2ye y 1
0
2
= 2ye y , y > 0
Consider,
(
x2 +y 2 )
f ( x ) . f ( y ) = 4 xy e = f ( x, y )
X and Y are independent RVs.
1
7 If the joint pdf of (X,Y) is f(x,y)= 4 ,0 x, y 2,
0, otherwise
Find P (x+y 1)
Solution :
1 1 y
P ( x + y 1) = f ( x, y )dxdy
0 0
TWO DIMENSIONAL RANDOM VARIABLES 17
1 1 y
1
= dxdy
0 0 4
1
1 1y
= [ x ]0 dy
40
1
1
4 0
= (1 y )dy
1
1 y2
= y
4 2 0
1 1
= 1 ( 0 0 )
4 2
11 1
= = .
42 8
Solution :
1 3
1 3
1
= ( 6 x y ) dxdy
0 2
8
3
11 x2
= 6 x xy dy
8 0 2 2
1
1 9
=
8 0
18 3 y 12 + 2 + 2 y dy
2
1
1 7
= y dy
8 02
1
1 7 y2
= y
8 2 2 0
1 7 1
=
8 2 2
3
=
8
(ii) P ( X + Y < 3 )
TWO DIMENSIONAL RANDOM VARIABLES 19
3 3y
= f ( x, y )dxdy
2 0
3 3y
1
= 8 ( 6 x y )dxdy
2 0
3 y
1
3
x2
= 6 x xy dy
82 2 0
1
2
3
= 6 ( 3 y )
( 3 y)
y ( 3 y ) dy
8 2 2
1
2
3
= 18 6 y
( 3 y)
3 y + y 2 dy
8 2 2
13 1
= 18 6 y 9 6 y + y 2 3 y + y 2 dy
82 2
3
1 6 y2 9 6 y2 y3 3y2 y3
= 18 y y+ +
8 2 2 4 6 2 3 2
1 54 27 54 27 27 27
= 54 + +
8 2 2 4 6 2 3
1 24 18 24 8 12 8
36 + +
8 2 2 4 6 2 3
5
=
24
P ( X < 1 Y < 3)
(iii) To find P ( X < 1 / Y < 3 ) =
P ( Y < 3)
TWO DIMENSIONAL RANDOM VARIABLES 20
2
1 x2
= 6 x xy
8 2 0
1
= [12 2 2 y ]
8
1
= (5 y )
4
3
P (Y < 3 ) = f ( y )dy
2
3
1
= ( 5 y )dy
2 4
3
1 y2
= 5 y
4 2 2
1 9
= 5 ( 3 2 ) + 2
4 2
1 9 5
= 7 =
4 2 8
3
P ( X < 1 Y < 3) 3
P ( X < 1 / Y < 3) = =8=
P ( Y < 3) 5 5
8
TWO DIMENSIONAL RANDOM VARIABLES 21
y
Find f and P (Y > 1 / 2 / X = 1 / 2 )
x
Solution
3
+ xy ,0 < x < 1, 0 < y < 1
Given f ( x, y ) = 4
0 , otherwise
y f ( x, y )
W.K.T. f = where f ( x ) is the marginal density of X.
x f ( x)
f ( x) = f ( x, y )dy
1
3
= + xy dy
0
4
1
3 xy 2
= y+
4 2 0
3 x
= + ,0 < x <1
4 2
3
+ xy
4 3 + 4 xy
y = =
Now f 3 x 3 + 2x
x +
4 2
1
y
P (Y > 1/ 2 / X = 1/ 2 ) = 1/ 2 f
x x=1/ 2
dy
TWO DIMENSIONAL RANDOM VARIABLES 22
1
3 + 4 xy
= 1/ 2 3 + 2 x x=1/ 2 dy
1
3+ 2y
=
1/ 2
3 +1
dy
1 1
= 3 y + y 2
4 1/ 2
1 3 1
= 4
4 2 4
1
= (16 6 1)
16
9
=
16
6
( x + y ) ;0 x 1, 0 y 1
2
f ( x, y ) = 5
0 , otherwise
1 3
Obtain the marginal PDF of X and that of Y. Hence find P y .
4 4
Solution :
6
( x + y ) ; 0 x 1, 0 y 1
2
Given that f ( x, y ) = 5
0 , otherwise
TWO DIMENSIONAL RANDOM VARIABLES 23
1
6
= ( x + y 2 ) dy
50
1
6 y3
= xy +
5 3 0
6 1
= x + ,0 < x 1
5 3
The marginal pdf of Y is
f ( y) = f ( x, y ) dx
1
6
=
50 ( x + y 2 ) dx
1
6 x2
= + xy 2
5 2 0
6 1
= y 2 + , 0 y 1.
5 2
3/ 4
1 3
P y = f ( y ) dy
4 4 1/ 4
6 1
3/ 4
5 y
2
= + dy
1/ 4 2
TWO DIMENSIONAL RANDOM VARIABLES 24
3/4
6 y3 y
= +
5 3 2 1/4
3/4
6 2 y3 + 3y
=
5 6 1 / 4
1 3/ 4
= 2 y 3 + 3 y
5 1/ 4
1 27 9 1 3
= 2 + 2 +
5 64 4 64 4
1 27 9 1 3
= +
5 32 4 32 4
1 26 6
= +
5 32 4
1 13 3
= +
5 16 2
1 13 + 24
=
5 16
1
= [37]
80
37
=
80
= 0.4625
TWO DIMENSIONAL RANDOM VARIABLES 25
e
( x + y )
, x 0, y 0
f ( x, y ) =
0 , otherwise
Find (i) P(X<1) and (ii) P(X+Y<1)
Solution
To find the marginal density function of X
Let the marginal density function of X be g(x) and it is defined as
g(x) = f ( x, y ) dy
= e
( x + y )
dy
0
e
x y
=e dy
0
= e x e y
0
= e x , x 0
g ( x ) = e x , x 0
1
P ( X < 1) = g ( x ) dx
0
1
= e x dx
0
= e x |10
= e 1 1
= 1 e 1
(
= e y 1 e (1 y ) dy
0
)
1
= e y e 1 dy
0
1
= e y e1 y
0
= e1 + e 1 1
= 1 2e1
TWO DIMENSIONAL RANDOM VARIABLES 27
y
(a) Evaluate C (b) Find fX(x) (c) fY / X and (d) fY(y).
x
Solution :
By the property of joint p.d.f, we have,
f ( x, y ) dxdy = 1
2 x
Cx ( x y ) dydx = 1
0 x
x
2
y2
C x xy dx = 1
0 2 x
2
2 x 2 2 x 2
C x x x dx = 1
0 2 2
2
x2 3x2
C x + dx = 1
0 2 2
2
C 2 x3 dx = 1
0
TWO DIMENSIONAL RANDOM VARIABLES 28
2
x4
2C = 1
4 0
16
2C = 1
4
1
C=
8
(b) The marginal density of X is given by,
x x
1
fX ( x) = f ( x) =
x
f ( x, y ) dy = x ( x y ) dy
8 x
x
x y2
= xy
8 2 x
x 2 x 2 2 x 2
= x x
8 2 2
x x2 3x2
= +
8 2 2
x
=
8
( 2 x2 )
x3
= ,0 < x < 2
4
1
x( x y)
y f ( x , y ) 8 1
f
(c) Y / X = = = ( x y), x < y < x
x f ( x) x3 2 x2
4
TWO DIMENSIONAL RANDOM VARIABLES 29
1 8
= 2 y
8 3
1
= (8 6 y )
24
1
= 2(4 3 y )
24
(4 3y)
= , x < y < x
12
TWO DIMENSIONAL RANDOM VARIABLES 30
EXERCISE
PART-A
I Choose the best answer
1FXY( , ) =
(a)1 (b)0 (c) (d)none
2E(XY) = .. if X and Y are independent
(a)XE(Y) (b)YE(X) (c)E(X)E(Y) (d)none
3 F(-,y) = .
(a)1 (b)0 (c) (d)-
Answers
(1)1 (2)E(X)E(Y) (3)0
PART-B
1For the following bivariate distribution , find (i) P(X 2 , Y = 2)
(ii)P(Y = 3) (iii)F(3) (iv)P(X < 3 , Y4)
X/Y 1 2 3 4
1 0.1 0 0.2 0.1
2 0.05 0.12 0.08 0.01
3 0.1 0.05 0.1 0.09
2 For the following bivariate distribution
X/Y 0 1 2
0 0.1 0.04 0.02
1 0.08 0.20 0.06
2 0.06 0.14 0.30
Compute the marginal pmfs, P(X 1 , Y 1) ,check if X and Y are independent.
3 The joint pdf of X and Y is given by
f(x,y) = (x+y)/21 , x = 1,2,3; y = 1,2.
Find the marginal distributions and conditional distributions.
4 The joint distribution function of X and Y is given by
F(x,y) = (1- e-x) (1- e-y) , x > 0,y > 0
0 , otherwise
(i)Find the marginal density function of X and Y.
(ii)Are X and Y independent?
(iii)Find the value of P[1 < X < 3 , 1 < Y < 2]
TWO DIMENSIONAL RANDOM VARIABLES 32
CHAPTER-II
density function f(x,y), then E[g(X, Y)] = g(x, y)f (x, y)dxdy
Properties
(i)E[g(X)] = g(x)f (x)dx, where f (x) is the marginal density of X.
x x
(ii)E[h(Y)] = h(y)f
y (y)dy, where f y (y) is the marginal density of Y.
Covariance
The covariance between two random variables X and Y is defined as
(
Cov(X, Y) = E X X Y Y
)( )
Correlation Coefficient
The correlation coefficient between two random variables X and Y is defined
Cov(X, Y)
as, xy =
x y
TWO DIMENSIONAL RANDOM VARIABLES 34
Note
(i)If xy = 0 then X and Y are said to be uncorrelated.
(ii)If E(XY) = 0 then X and Y are said to be orthogonal.
Note
V(X + Y) = V(X) + V(Y) + 2 Cov(X ,Y)
Properties of covariance
Cov (aX , bY) = ab Cov (X,Y)
Cov(X + a ,Y + b) = Cov(X , Y)
Var(aX + bY + c) = a2 Var(X) + b2 Var(Y) + 2ab Cov(X , Y)
Cov(X ,Y) = E(XY) E(X) E(Y)
Properties of correlation coefficient
-1 (X , Y) 1
Correlation coefficient is independent of change of origin and scale. i.e., If U = (X-a)/h and V
= (Y-b)/k where h , k > 0 then
(X , Y) = (U , V).
Problems
1 Suppose the joint propbability mass function of a RV (X,Y) is given by,
1
, for (0,1), (1, 0), (2, 2)
= 3
0, otherwise
Y / X 0 1 2 P(y) =
0 0 1/3 0 1/3
1 1/3 0 0 1/3
2 0 0 1/3 1/3
P(x) = 1/3 1/3 1/3 1
TWO DIMENSIONAL RANDOM VARIABLES 35
We have =
= and =
= . =
V(X) = E(
Var(Y) = E(
E (XY) =
2 If X ,Y and Z are uncorrelated RVs with zero mean and S.D. 5 , 12 and 9
respectively, and if U = X + Y and V = Y + Z , find the correlation coefficient between U
and V.
Solution.
Since X, Y and Z are uncorrelated,
Cov (X,Y) = 0 ; Cov (Y,Z) = 0 and Cov (Z,X) = 0
i.e., E(XY) E(X) E(Y) = 0;
E(YZ) E(Y) E(Z) = 0 and
E(ZX) E(Z) E(X) = 0
Given E(X) = 0 = E(Y) = E(Z)
Also given S.D.of X = 5;
S.D.of Y = 12
and S.D.of Z = 9
i.e., Var X = 25 , Var Y = 144 and Var Z = 81
E = 25; E( = 144 and E(
(Q Var X = E ( X ) [ E ( X )] )
2 2
To find :
Var U
= Var (X+Y)
= Var X + Var Y + 2 Cov (X,Y)
= 25 + 144 + 0 = 169
= 13
Var V
= Var (Y+Z)
= Var Y + Var Z + 2 Cov (Y,Z)
= 144 + 81 + 0
= 225
= 15
= = 0.7385
X: 65 66 67 67 68 69 70 72
Y: 67 68 65 68 72 72 69 71
TWO DIMENSIONAL RANDOM VARIABLES 38
Solution :
X Y XY
Y=
Y = 552 = 69
n 8
= 68 69 = 4692
Cov (X,Y) =
= 4695 4692 = 3
The correlation Co-efficient of X and Y is given by,
TWO DIMENSIONAL RANDOM VARIABLES 39
r (X, Y) or (X , Y) =
Solution :
The marginal density function of X is
f ( x) =
f ( x, y )dy
1
= (2 x y ) dy
0
1
y2
= ( 2 x ) y
2 0
1
= 2 x
2
3
= x, 0 < x < 1.
2
Similarly, the marginal density function of Y is,
1
f ( y) = f ( x, y ) dx = ( 2 x y ) dx
0
TWO DIMENSIONAL RANDOM VARIABLES 40
1
x2
= ( 2 y ) x
2 0
1
=2 y
2
3
= y, 0 < y < 1
2
E (X) =
xf ( x ) dx
1
3
= x x dx
0 2
1
3
= x x 2 dx
02
1
3x 2 x 3
=
4 3 0
3 1
=
4 3
94
=
12
5
=
12
5
Similarly, E (Y) =
12
TWO DIMENSIONAL RANDOM VARIABLES 41
x f ( x ) dx
2 2
E( X ) =
1
23
= 0 2 x dx
x
1
3
= x 2 x 3 dx
0
2
1
x3 x 4
=
2 4 0
1 1
=
2 4
1
=
4
1
Similarly E( =
4
1 5
2
=
4 12
TWO DIMENSIONAL RANDOM VARIABLES 42
1 25
=
4 144
36 25
=
144
11
=
144
11
Similarly, Var (Y) =
144
1 1
E ( XY ) = xyf ( x, y ) dxdy
0 0
1 1
= xy ( 2 x y ) dxdy
0 0
1 1
= ( 2 xy x 2 y xy 2 ) dxdy
0 0
1
1
x3 y x 2 y 2
= x2 y dy
0
3 2 0
1
y y2
= y dy
0
3 2
TWO DIMENSIONAL RANDOM VARIABLES 43
2 y y2
1
= dy
0 3 2
1
y 2 y3
=
3 6 0
1 1
=
3 6
2 1
=
6
1
=
6
Cov (X,Y) = E(XY) E(X) . E(Y)
1 5 5
= .
6 12 12
1 25
=
6 144
24 25
=
144
1
=
144
The correlation Co-efficient is
TWO DIMENSIONAL RANDOM VARIABLES 44
Cov ( X,Y )
xy =
x y
-1
= 144
11 11
12 12
-1
=
11
Solution :
The marginal density function of X is given by,
f ( x) =
f ( x, y )dy
1
= ( x + y ) dy
0
1
y2
= xy +
2 0
TWO DIMENSIONAL RANDOM VARIABLES 45
1
= x + ,0 < x < 1
2
The marginal density function of Y is given by,
f ( y) = f ( x, y )dx
= ( x + y ) dx
0
1
x2
= + xy
2 0
1
= y + ,0 < y <1
2
1
E ( X ) = xf ( x ) dx
0
1
1
= x x + dx
0 2
1
x
= x 2 + dx
0 2
1
x3 x2
= +
3 4 0
1 1
= +
3 4
7
=
12
TWO DIMENSIONAL RANDOM VARIABLES 46
1 1
1
E (Y ) = yf ( y ) dy = y y + dy
0
0
2
1
y3 y 2 7
= + =
3 4 0 12
1
1
1
1
x2
E ( X ) = x f ( x ) dx = x x + dx = x3 + dx
2 2 2
0 0 2 0
2
1
x 4 x3 1 1 3+ 2 5
= + = + = =
4 6 0 4 6 12 12
5
||| ly E (Y 2 ) =
12
2
V ( X ) = E ( X 2 ) E ( X )
5 49
=
12 144
11
=
144
11 11
X = and ||| ly Y =
12 12
TWO DIMENSIONAL RANDOM VARIABLES 47
1 1
E ( XY ) = ( xy )( x + y ) dxdy
0 0
1 1
= ( x 2 y + xy 2 ) dxdy
0 0
1
1
x3 y x 2 y 2
= + dy
0 0
3 2
y y2
1
= + dy
0
3 2
1
y 2 y3
= +
6 6 0
1 1 2 1
= + = =
6 6 6 3
TWO DIMENSIONAL RANDOM VARIABLES 48
Cov ( X , Y ) = E ( XY ) E ( X )(Y )
1 7 7
= .
3 12 12
1 49
=
3 144
48 49
=
144
1
=
144
1
Cov ( X , Y ) 1
rXY = = 144 = = 0.0909
X Y 11 11 11
.
12 12
3 ( x + y ) , 0 < x 1,0 y 1, x + y 1
f(x,y) =
0 , otherwise
Find (1) the marginal p.d.f. of X
(2) P[X + Y < ]
(3) cov(X,Y)
TWO DIMENSIONAL RANDOM VARIABLES 49
Solution :
Given
3 ( x + y ) ,0 < x 1, 0 y 1, x + y 1
f (x,y) =
0 , otherwise
g(x) =
=3
=3
=3
=3
P[X + Y < ] =
=3
= 3 dx
=3
TWO DIMENSIONAL RANDOM VARIABLES 50
= dx
E[XY] =
=3
=3
=3
TWO DIMENSIONAL RANDOM VARIABLES 51
=3
=3 dx
= dx
= dx
=
TWO DIMENSIONAL RANDOM VARIABLES 52
E[x] =
Similarly E(y) =
= -13/320
TWO DIMENSIONAL RANDOM VARIABLES 53
X
5 15
Y
10 0.2 0.4
20 0.3 0.1
Solution :
X
5 15 p(y)
Y
10 0.2 0.4 0.6
20 0.3 0.1 0.4
p(x) 0.5 0.5 1
E[X] =
E[Y] =
E[XY] =
= 50
Regression
Regression is a mathematical measure of the average relationship between two or
more variables interms of the original limits of the data.
Lines of Regression
If the variables in a bivariate distribution are related we will find that the points in
the scattered diagram will cluster around some curve called the curve. If regression of the
curve is a straight line, it is called the line of regression between the variables, otherwise
regression is said to be curvilinear.
The line of regression of y on x is given by,
y
y y = r.
x
( x x)
x
x x = r.
y
(
yy )
Where r is the correlation coefficient and x , y are standard deviations.
Note
Both the lines of regression pass through ( X, Y ).
Regression Coefficients
Regression coefficient of y on x is
y
b yx = r
x
Regression coefficient of y on x is
TWO DIMENSIONAL RANDOM VARIABLES 55
x
b xy = r
y
Note
Correlation coefficient r = b xy b yx
1 r 2 x y
tan =
r x 2 + y 2
Note
When there is no linear regression between X and Y then the two lines of regression
are at right angles
i.e., if r = 0 then tan = 1 =
2
Problems on Regression
1 The regression lines between two random variables X and Y is given by 3X + Y = 10
and 3X + 4Y = 12. Find the co-efficient of correlation between X and Y.
Solution :
Assume that 3X + Y = 10 be the regression line of X and Y.
3X = -Y + 10
4Y = -3X + 12
5 = 22 (1)
64 = 24 (2)
45 (3)
(3) (2) =6
(1)
X Y X=x 69 Y = y-152 XY
3 Find the coefficient of correlation & obtain the lines of regressions from the data
given below :
x: 62 64 65 79 70 71 72 74
Cov (X,Y) =
TWO DIMENSIONAL RANDOM VARIABLES 58
= 95.125
Y- = r.
Y- 155 = 0.9
= 6.02
Y = 6.02X 411.62 + 155
Y = 6.02X 256.62
Similarly the regression line of X on Y is,
X-
X 68.375 = 0.9
= 0.13
X = 0.13Y + 68.375 20.15
X = 0.13Y + 48.23
TWO DIMENSIONAL RANDOM VARIABLES 59
EXERCISE
PART-A
I Choose the best answer
1The random variables are uncorrelated with each other, if the correlation between them
is equal to ..
(a) (b)1 (c)0 (d)none
2The correlation coefficient ranges from to ..
(a)0,1 (b)-1,1(c)-1,0 (d)none
3 The correlation coefficient = .
(a)cov(x,y)/x2 y2 (b) cov(x,y)/2x2 y2 (c) cov(x,y)/x y (d)none
4 The angle between two regression lines is given by tan
(1 r 2 ) x y (1 r 2 ) x y (1 r 2 ) x 2 y 2
(a) X (b) X (c) X 2
r x 2 + y2 r2 x 2 + y2 r x + y 2
(d)none of these
Answers
(1 r 2 )
(1)0 (2)-1,1 (3) cov(x,y)/x y (4) X 2x y 2
r x + y
PART-B
1 Derive the limits of regresion.
2 Derive the angle between two lines of regression.
3 Find the correlation coefficient for the following data
X 42 44 58 55 89 98 66
Y 56 49 53 58 67 76 58
4 The joint pdf of X and Y is given by
f(x,y) = cx2(1- y) , 0 x,y 1
Find the correlation coefficient between X and Y.
5 For the following bivariate distribution
X/Y 0 1 2
0 0.1 0.04 0.02
1 0.08 0.20 0.06
2 0.06 0.14 0.30
Find the Correlation coefficient between X and Y.
6 Obtain the Regression lines for the following data
TWO DIMENSIONAL RANDOM VARIABLES 61
X 1 2 3 4 5 6 7
Y 9 8 10 12 11 13 14
CHAPTER-III
TRANSFORMATION OF RANDOM VARIABLES
INTRODUCTION
Most of the times, we subject the signals to undergo transformation, in order to derive
certain advantages of the signal. For example, a weak signal received may be amplified in
order to increase the amplitude of the signal. An ac signal may be rectified to yield a dc
signal. If the input is a random variable with a particular probability density or distribution
function and when it undergoes a transformation yielding another random variable; a
natural question arises as to what the probability density or distribution function of the
output random variable is. In other words, what is the effect of transformation unit on the
pdf of input random variable?
In general, transformation may be of different nature. A single random variable may be
transformed into another single random variable or more than one random variables may
be transformed into a single random variable or multiple random variables may be
transformed into another set of multiple random variables.
x x
(x, y) u v
y)J where J = =
(u, v) y y
u v
U+V
X=
2
U V = 2Y
UV
Y=
2
x x
(x, y) u v
J= =
(u, v) y y
u v
1 1
2 2
=
1 1
-
2 2
1 1
=
4 4
1
=
16
TWO DIMENSIONAL RANDOM VARIABLES 64
g(u, v) = J f (x, y)
1 u+v uv
= f ,
16 2 2
Y
R = X 2 + Y 2 and = tan 1 .
X
Solution
Since X and Y are independent normal random variables
with mean zero and variance 2,
(x 0) 2
1
22
f (x) = e , < x < and
2
(y 0) 2
1
22
f (y) = e , < y <
2
x2
1 2
f (x) = e 2 , < x < and
2
y2
1 2
f (y) = e 2 , < y <
2
Since X and Y are independent,
f(x , y) = f(x) f(y)
x2 + y2
1
2 2
= 2
e , < x, y <
2
TWO DIMENSIONAL RANDOM VARIABLES 65
Y
Here R = X 2 + Y 2 and = tan 1
X
i.e., R 2 = X 2 + Y 2
X = R cos , Y = R sin
x x
r
J =
y y
r
cos -rsin
=
sin rcos
= r cos2 + r sin 2
=r
g(r, ) = f (x, y) J
( x2 + y 2 )
1 2 2
= e .r
22
r2
1 2 2
= e .r , 0 < r < , 0 < < 2
22
TWO DIMENSIONAL RANDOM VARIABLES 66
g R (r) = g(r, )d
2 r2
1 2 2
=
0
2 2
e .rd
r2
r 2
= 2
e 2
[ ]20
2
r2
r 2
= e 2
.2
2 2
r2
r 2
= 2 e 2 , 0 < r <
g () = g(r, )dr
r2
1 2 2
= 2
e .rdr
0
2
1 t 2 r2
= e dt [put 2
= t 2rdr = 22 dt]
0
2 2 2
TWO DIMENSIONAL RANDOM VARIABLES 67
1 e t
=
2 1 0
1 0
= - e e
2
1
=- [ 0 1]
2
1
=
2
1 u
-
= v v2
0 1
1 1
= 0 =
v v
TWO DIMENSIONAL RANDOM VARIABLES 68
g(u, v) = f (x, y) J
1
= (x+y).
v
u 1
= + v .
v v
u
= +1 , 0 u v , 0 v 1
v2
u
[Q 0 x 1 0 1 0 u v]
v
[and 0 y 1 0 v 1]
v
v=1 u=v
u=0
v=0 u
g u (u) = g(u, v)dv
1
u
= 2 + 1 dv
u
v
1
u
= + v
v u
= -u+1-(-1+u)
= -u+1+1-u
= 2-2u
x x
u v
J=
y y
u v
1 1
=
0 1
= 1-0 = 1
The joint pdf of (U , V) is given by,
g(u, v) = f (x, y) J
= e (x + y) .1
[v > 0 and u < v]
(u + v + v)
=e
= e (u + 2v)
The pdf of U is given by,
TWO DIMENSIONAL RANDOM VARIABLES 71
g u (u) = g(u, v)dv
e
( u + 2v)
= dv
u
e
u 2v
=e dv
u
e 2v
u
=e
2 u
e2u
= e u 0 +
2
eu
= ,u < 0
2
TWO DIMENSIONAL RANDOM VARIABLES 72
EXERCISE
PART-A
I Choose the best answer
1If f(x) = 3x and Y = 9X, then the pdf of Y is ..
(a) x/4 (b)x/3 (c)x/5 (d)none
2 If f(x) = e-x and Y = X1/2, then the pdf of Y = .
y2
(a) 2ye (b) 2ye y (c)2yey (d)none
3 If x = uv and y = v then J = .
(a)1 (b)1-v (c)v (d)none
4 If x = v and y = u/v then J =
(a) 1/v (b)-1/v (c)v (d)none
Answers
y2
(1)x/3 (2) 2ye (3)v (4)-1/v
Answers
1 z
Answers
(1)False (2)True
PART-B
1 Find the pdf of Y = eX where X is a normal random variable with mean 0 and variance 2.
2 If X is a normal random variable with mean zero and variance 2, find the pdf of Y = X1/2.
3 If X is exponentially distributed with parameter a, find the pdf of Y = log X.
4 Let (X,Y) be a two dimensional random variable having the density function
2
+ y2 )
f (x, y) = 4xye(x , x 0 ,y 0
Find the density function of U = (X2 + Y2).
5 The joint pdf of two rvs is given by