PROBLEMS ON TWO DIMENSIONAL RANDOM VARIABLE

1.State var (x) = E(x2) – E(x)2

Proof

Var ( x)  E  x   x 
2

=E  x 2  2  x X   x 2 
=E  x 2   2  x E ( x)   x 2   x is a constant 
=E  x 2    x 2   x  E ( x) 
=E  x 2   [ E ( x) 2 ]

2. If X is a discrete or continuous RV P.T E(ax+b) = aE(x)+b & Var (ax) = a2var(x)

Proof

E  ax  b   E  ax  b 
  a x i pi  b
j

 a  xi pi  b  aE ( x)  b
j

Var (ax)  E  a 2 x 2    E (ax) 2

=a 2 E ( x 2 )  ( aE ( x) 2
=a 2 [ E ( x 2 )  [ E ( x) 2 ]]
=a 2 var( x)
3.The joint p.d.f of X and Y is given by
f(x,y)=3(x+y), 0x1, 0y1. Find marginal density function of X and Y.

Solution:

Marginal density of X:

1 1
fx (x)   3(x  y).dy  3  xy  y 2 
 2 0
0

=3  x+ 1 
 2

Marginal density fun of y:

 1 1
fy (y)   3(x  y).dx  3  x 2  xy 
 2 0
0

=3  1  y
 2 
4.For the bivariate probability distribution of (x,y) given below find p(x <1) ,p(y<3),

p( x  1, y  3), p(x  1/y  3), p(y  3/x  1)

and p(x+y  4)
x/y 1 2 3 4 5 6
0 0 0 1 2 2 3
32 32 32 32
1 1 1 1 1 1 1
16 16 8 8 8 8
2 1 1 1 1 0 2
32 32 64 64 64
Solution:

P(x  1 ) = P (x = 0) + p(x =1)

6 6

 p( x  0, y=j) + P( x  1, y  j )
j 1 j 1

1 2 2 3 1 1 1 1 1 1
(0  0 
   )     
32 32 32 32 16 16 8 8 8 8
1 5 7
  
4 8 8

p ( y  3)  p  y  1  p ( y  2)  p ( y  3)
2 3 2
  p ( x  i, y  1) + p  x=i. y=2    p (x=i. y=2)
i 0 i=0 i 0

 1 1   1 1   1 1 1 
 0   0      
 16 32   16 32   32 8 64 
3 3 11 23
   
32 32 64 64
 3 3
P ( x  1, y  3)   p ( x  0, y  j )   p ( x  1, y=j)
j 1 j 1

 1   1 1 1 9
=  0+0+       
 32   16 16 8  32
p ( x  1, y  3) 9 / 32 18
P ( x  y  3)   
p(y  3) 23 / 64 23
p ( x  1, y  3) 9/32 9
p ( y  3 / x  1)  = 
p(x  1) 7/8 28
4 3
p( x  y  4)   P( x  0, y  j )   p ( x  1 y=j)
j 1 j 1
2
+ p(x=2, y=j)
j=1

3 1 1 13
   
32 4 16 32

5.The joint pdf of the RV(x,y) is given by f(x,y) =kxy e-(x2+y2) x > 0 , y>0 Find the value
of k prove also that X & Y are independent.

Solution:

Here the range space is the entire first quadrant of the xy-plane.

By the property of the joint pdf


 x2  y 2  dxdy  1
  kxye
x 0 y 0
 
(ie) K   Ye  y dy  xe-x dx  1
2 2

0 0 0

k
1 k  4
4

 x2 -y 2
now f x (x)=  4 x e .e .dy  2 xe  x
2

Similarly

f y ( y)  2 ye y , y  0
2

Now f x (x)  f y (x)=4xye-(x  f ( x, y)

2
+y2 )
The RVs X & y are independent.

6.The joint distribution function of 2 random variables X and Y is given by

F(x,y)=1-e-x-e-y+e-(x+y), x>0,y>0.

Find the density function of x and y.

Solution:

2
The joint p.d.f f(x,y)= F(x,y)
xy

2 
1  e x  e y  e   
 xy

xy  

= e y  e  x  y  
x  
=e 
- x+y 

The joint p.d.f of r.v X and Y is given by

 1 yx  y
 .e .e , x > 0, y > 0
f (x, y)   y
0
 , otherwise

7.The joint pdf of the RV(x,y) is given by f(x,y) =kxy e-(x2+y2) x > 0 , y>0 Find the value
of k prove also that X & Y are independent.

Solution:

Here the range space is the entire first quadrant of the xy-plane.

By the property of the joint pdf

 
 x2  y 2  dxdy  1
 
x 0 y 0
kxye

 
(ie) K   Ye  y2
dy  xe-x dx  1
2

0 0 0

k
1 k  4
4

 x2 -y 2
now f x (x)=  4 x e .e .dy  2 xe  x
2

Similarly

f y ( y)  2 ye y , y  0
2

Now f x (x)  f y (x)=4xye-(x  f ( x, y)

2
+y2 )

The RVs X & y are independent.

8. X and Y are two R.V. with joint pdf

1
f(x,y)= (6-x-y), 0<x<2, 2<y<4
8
find (1) P(x<1y<3) (ii) P(x+y<3)
(iii) P(x<1/y<3)

Solution:
 1 3 1 3
1
(i ) P  x  /  y  3    f ( x, y ) dydx     6  x  y  dydx.
0 2 0 2
8
3
1  y2  1 7 
1 1
 
80
 6 y  xy  
2 2
dx   
8 0 2
 x  dx

1
1 7 x2  1
  x    6  3
8 2 2  0 16 8

2 3 x
(ii ) P( x  y  3)    f ( x, y )dydx
0 2

2 3 x 3 x
1 1 y2 
2
1
  (6  x  y )dydx    6 y  xy   dx.
0 2
8 80 8 2 2
2
1 7 x2  1 7 4 x 2 x3 
2
    4 x   dx   x   
8 0 2 2 8 2 2 3 0

1 2 1
  
8 6 24
P( x  1  y  3)
P  x  1/ y  3  (1) 
P( y  3)
2 2
1
f ( y ) ( y )  Marginal pdf of y= 
0
f ( x, y )dx   (6  x  y )dx
0
8

1
= 10  2 y 
8
3 3
1
P(Y  3)  f
2
y ( y )dy   (10  2 y )dy
2
8
3
1 2 y2 
 10 y    8
5
8 2 2
(1)
1 3  by 
P  x |  y  3   f xy ( x, y )dxdy  3  
8 
2 2 
3
P( x  /  y  3)
P( x  1/ y  3)   83
P( y  3) 5 5
8

9. Two R.V. x and y have the joint pdf

2  x  y 0<x<1,0<y<1
f ( x, y )  
 0 otherwise
S.T. correlation r(x,y)=-1/11

Solution:
 1 1
f x ( x)   f xy  x, y  .dy    2  x  y  .dy 
0 0
1
  2
  2 y  xy  y   3  x, 0  x  1
 2 0 2
1
1 1
 x2 
f y ( y )   f xy  x, y  .dy    2  x  y  dx   2 x   y x   3  y
2
0 0  2 0

 f y ( y )  3  y, 0  y  1
2

 
1 1
E ( x)   x. f x  x  .dx   x 3  x dx
2
0 0

1
 x 2 x3 
 3  5
2 2 3 12
 0
1


y 2 y3 

1 1
E ( y )   y. f y  y  .dy   y 3  y dy   3  
2 2 2
0 0  3 0
1

 x3 x 4 

1 1
E ( x 2 )   x 2 . f x  x  .dx   x 2 3  x dx   3 
2 2 2 3
0 0  0
1 1
  1
2 4 4

 
1 1
E ( y 2 )   y 2 . f y  y  .dy   y 2 3  y dy
2
0 0

1
 y3 y 4 
 3   1
2 3 4
 4 0
1 1
E ( xy )    xy. f  x, y  .dxdy
0 0
1
1 1
 y2 y2 y3 
1
   xy(2  x  y)dxdy   2x  x2  x  dx
0
0 0
2 2 3 0
1

1
x2 x   x2 x3  1
  x    .dx  2    
0
2 3  6 6 0 6

Cov(x,y)  E(xy)  E(x).E(y)

1 5 5 1
   
6 12 12 144
V(x)  E(x 2 )  E  x  
2

2
1 5 11
   
4 12  144
11
V(y)  E(y 2 )  E(y) 
2

144
1

Cov(x,y) 144 1
   
V(x). V(y) 11 11 144

144 144 11
144
1

11

10. Find the regression lines:

X: 6 8 10 18 20 23
Y: 40 36 20 14 10 2

Solution:

Regression equation of x on y:

x  x  bxy (y  y)
The regression equation of y on x is
y  y  byx (x  x)

Total

x Y x2 y2 xy
6 40 36 1600 240
8 36 64 1296 288
10 20 100 400 200
18 14 324 196 252
20 10 400 100 200
23 2 529 4 46
85 122 1453 3596 1226
x
 x  85  14.17
n 6

y
 y  122  20.33
n 6

x  x 
2 2 2
1453  85 
x         6.44
n  n  6  6 

y
y 2
y
  
2
3596  122 

2

  13.63
n n 6  6 

 xy  xy 1226
 (14.17)(20.33)
 n  6  0.95
x. y (6.44)(13.63)

x 6.44
b xy  .  0.95   0.45
y 13.63

y 13.63
byx  .  0.95   2.01
x 6.44

The regression equation of x on y is

x  x  bxy (y  y)
 x  14.17  0.45 (y  y)
 x  0.45y  23.32

The regression equation of y on x is

y  y  b yx (x  x)
y  20.33  2.01 (x  14.17)
 y  2.01 x  48.81

1 kx  x  y  , 0  x  2, |y |  x
11.The joint p.d.f of 2r.v. X and y is f  x,y    Evaluate the

0 , otherwise
constant K. Find the marginal and conditional density of y given x.
Solution:
2 x

 
(i)  k x 2  xy dydx  1
0 x

x
2
 2 y2 
K   x y  x  dx  1
0
2 x

  
2 x
 K   x 3 y  x 3   x 3  x 3  dx  1
 2 2   x
0
2 2
 4 
 K  2x 3 dx  1  K  x 1
0  4  0
K 4
 2  0   1
2

8k=1  K= 1
8

(ii) Marginal density of X and

x x
 
 
2
f  x   1  x 2  xy dy  1  x 2 y  xy 
8 8 2 x
x

  x 3  x3   x 3  x3 
1
8  2 
2  
1
  2x 3  x 3
8 4

The marginal density of y

 2  2

  1
 
g  y   1   x 2  xy dx  where -2<y<0 and g(y)=  x 2  xy dx where0  x  2
8
  y  8 y
1 3
     1  8 y3 y3 
2 2 2
g(y)   x y x    2y  
8 3 y 2  y  8  3 3 2
1 y 5y 3
   in  2  y  0
3 4 48
and

   2  
1  x3 2 2
g y 
2
-y x
8  3 y y

1 8 y 2  2y  y 3 
 
8  3 3 2 
1 y y3
   in0  y  2
3 4 48

 The marginal density of Y is

1 y 5y 3
  4  2 y  0
3 48
g(y)   3
 1  y  y in 0  y  2
 3 4 48

e( x  y) ,0  x,y  
12. If f(x,y)  is the joint p.d.f of a random variable X and Y, find (i) p(x<).
0 , Otherwise

(ii) p(x>y) (iii)P(x+y<1)

Solution:
 1 1
 x  y  
P(x  1)    e dydx   e x  e  y  dx
0
0 0 0
0
1
 1 e x dx  1 

e
 x
 x  y 
P(x  y)    e dydx
0 0

x
  e x  e y  dx
0
0

 
  e x 1  e x  .dx    e x  e2x  .dx
o 0 

  x e2x    1 
  e    0   1   
 2 0   2 
1

2

1 1-x

iii p  x+y<12 e

- x+y 
.dy dx.
0 0
1 1
 y 1 x 1 x 
  e  e 
x
dx=-  e x e  1 .dx.
0  
0 0
1 1
= e x .dx  e1  dx
0 0

2
  1 x 0
1 1

= -e-x e 1 .
0 e

4ax 0  x  r
13. The independent random variables X and Y are defined by f (x)  
0 otherwise
4by 0  y  s
f (x)   find the correlation co-efficient between U=x+y and V=x-y.
0 otherwise

Solution:

To find a and b
 r s

 f  x .dx  1.
0
&  f  y .dy  1
o
r s
 4a  x.dx  1.  4b  y.dy  1.
o o

1
 2ar 2  1 2bs 2  1  b=
2s2
1
a .
2r 2
2x 2y
 f  x   2 , 0  x  y and f  y   2 , 0  y  s.
r s
since x and y are independent

r  x,y   0  cou  x,y   0

cov  u,v   cov  x  y, x-y    x2   y2

v  u   v  x  y    x2   y2 .
v  v   v  x  y    x2   y2 .
cov  u,v 
 r  u,v  
 u . v
 x2   y2
= 2
 x   y2
r r
2 2 2r
Now E  x    x.f  x  .dx  2 
x .dx  .
0
r 0 3
r r
r2
E x  2 2 2
  x .f  x  .dx  2  x 3 .dx  .
r 0 2
0

r 2 4r 2
 
 v  x   E x 2   E  x    
2

2 9
2
r 1
= 
18 36a

1
III1Y v  y  
36b
1 1

ba
Hence r u,v   36a 36b  .
1

1 b  a
36a 36b
14. The equations of two regression lines are 3x+12y=19, 3y+9x=46 obtain.
(i) the correlation co-efficient (ii) mean value of x and y.

Solution:

Let regression line of y on x:3x+12y=19

3 19
y x .
12 12

Let regression line of x on y: 3x+9y=46

1 46
x y .
3 9
3 1
 b yx  and b xy  .
12 3
3 1 1
 r 2  b yx .b xy    .
12 3 12
1
r 
2 3
-1
 r= .
2 3

(ii) Both regression lines passes through the point

 x,y . 1
3x  12y  19  (1) and 9x+3y  46  (2) solve (1) & (2) we get x  5, y  .
3

1 1
15.Let the random variables x have the marginal density f(x)=1,  x  and let the
2 2
 -1
 1, x  y  x  1, x0
y  2
conditional density of y given x be f    
 x  1, -x<y<1-x, 0<x<
1
 2
Show that the variables x and y are uncorrelated.

Solution:
 1 1 1
2
 x2  2
2
E  x    x.f  x  .dx   x.dx    .
1 1  2  1
2 2 2

1  1 1
=     0.
2 4 4
If f(x,y) is the joint put f of x and y, then

y y
f  x,y   f  x  . f    f   .
x x
1
0 x 1 2 1 x
E  xy     xy1.dy dx    xy dy dx.
1 x 0 x
2

1 1 x
x 1
1
0
y  2
1 y  2 2
 =  x   dx   x.   .dx.
2 1  2  x 2 0  2 x
2
1
6 2
1 1

2 1 x  2x  1. dx  2  x (1  2x xdx).
0

2
1
6
1  2x 3 x 2  1  x 2 2x 3  2
       
2 3 2  1 2  2 3 0
2

1  1 1 1 1
     0.
2 12 8 12 8 
 Cov  x1y   E  xy   E  x  . E  y   0.
The variables X and Y are uncorrelated.