Fayoum University Applied Statistics

Faculty of Engineering Second Term of

2008/2009
Dept. of Electrical Engineering (Communication and Electronics Section)
Final Exam
Second Year Time: 3 hours
Model Answer

1) p̂ =2/3 , n=1600

a) The (1-  ) confidence interval for p is

pˆ (1  pˆ )
pˆ  z / 2
n

(1-  )=0.95   =.05   / 2 =0.025 and z / 2  z 0.025  1.96

The 95 % confidence interval for p is

2
 1.96
1  2  1   2  1.96 2  2  0.0230988
  
3 1600  3  3  3 40  3 3

= ( 0.643568 , 0.689765 )

2
pˆ (1  pˆ ) z 
b) The precision E  z / 2  n  pˆ (1  pˆ )   / 2 
n  E 
2 2
1z   1.96 
g ( pˆ )  pˆ (1  pˆ )  g max  g (0.5)  0.25  n    / 2  = 0.25    = 2401
4 E   0.02 

2) The subscript 1 denotes cotton fiber

The subscript 2 denotes acetate

n1  25 , s1  1.5 , x1  20
 n 2  25 , s 2  1.25 , x 2  12

y  x1  x2 , Y  X1  X 2 , Y  X1  X 2

E (Y )  E X 1   E X 2   E X 1   E X 2   1   2

 12  22  
 Y2  var(Y )  var( X 1  X 2 )  var( X 1 )  var( X 2 ) =  =  2  1  1 
n1 n2  n1 n2 

( n1  1) s12  ( n2  1) s22 2 2
sY  s 1

1
where s 2   s 2  (24)(1.5)  ( 24)(1.25)
n1 n2 (n1  n2  2) 48

1
s 2  (1.5) 2  1.25) 2  =1.90625= (1.380670) 2
2 
H0
sY  1.380670 1  1 = 0.390512
25 25
HA
H0 : y  0

H A : y  0

C0
 tn
1  n2  2 ,
C  sY t n
sY 1
n
2
 2, 

C
t t z  1.645
n  n  2,  48, 0.05 0.05
1 2

C  (0.390512)  (1.645)  .642393

y  x1  x 2 =20-12=8  y C

Decision : We reject H 0  We accept H A

There is strong evidence that 1   2

3) n=5

1 n 1 n
a) x   xi  3 , s2   ( xi  x ) 2  0.815  (0.9027735) 2
n i 1 ( n  1) i 1
  
2  (n  1) s 2 (n  1) s 2 
b) The (1-  ) confidence interval for  is  , 
  n2  1,  / 2  n2  1, 1   / 2 
 

(1-  )=0.95   =.05   / 2 =0.025

2 =
2
=11.143 and 2 =
2
=0.484
n  1,  / 2 4, 0.025 n  1, 1   / 2 4, 0.975

The 95 % confidence interval for  2 is

f (x )
 4 * 0.815 4 * 0.815  = ( 0.292560 , 6.735537 )
 , 
 11.143 0.484 
HA
C s
0
c)   t n  1,   C
0
t
n  1, 
H0
s/ n n

t =t =1.533 
n  1,  4,0.1

0.815 = 2.881
C  3.5  1.533
5 A C 0 x
x 3  x C

Decision : We accept H 0 .

f ( 2 )
d) The decision criteria :

 02 n2  1, 1   / 2  02 n2  1,  / 2
C1= and C2= 
n 1 n 1
2 
 =.05   / 2 =0.025
2

2 =
2
=11.143
n  1,  / 2 4, 0.025
2 2 2
n  1, 1   / 2 n  1,  / 2

2 =
2
=0.484
n  1, 1   / 2 4, 0.975

(1)  (0.484) (1)  (11.143)

C1= = 0.121 , C2 = = 2.78575
4 4

s 2  0.815  C1  s 2  C2  Decision : We accept H 0 .

4) Ai : The event that engine i is available when needed.

P( Ai )  0.96 i =1, 2

A1 and A2 are statistically independent.

a) P( A1  A2 )  P( A1 )  P( A2 )  (1  0.96) 2  (0.04) 2  0.0016

b) Pb  P( A1  A2 )  P ( A1 )  P ( A2 )  P( A1  A2 )  P( A1 )  P( A2 )  P( A1 ) * P ( A2 )

 0.96  0.96  (0.96) 2  0.9984

5) Success: getting a head in one toss

p=0.45  q=1-p=1-0.45=0.55
th
X : number of failures before the r success  X follows the negative

binomial distribution

p ( x ) C pr q x
x  r 1 x
, x  0 , 1, 2 , 3, ....... . .
r = 4, x+r = 11  x = 11-4 = 7
p(7)  10
C7 (0.45)4 (0.55)7 p(7)  0.0749152

6)
x 1 2 3 4 5 6

p(x) 1/6 1/6 1/6 1/6 1/6 1/6

 6
1 6 21
 x  E X    x p ( x)   x   3.5
x 1 6 x1 6
6
1 6 2 91
 
E X 2   x 2 p( x)  
6 x 1
x 
6
 15.166667
x 1

 X2  E ( X 2 )   X2  2.916667

X and Y are identically distributed

Z  X  3Y  5

 z  EZ   E X   3E Y   5

Z   z   X  EX   3Y  EY 

Z   z 2   X  E X 2  9Y  EY   6 X  E X Y  E X 

2

 2
  2
  2

E Z   z   E  X  E X   9 E Y  E Y   6 E  X  E X Y  EY 

 Z2   X2  9 Y2  6 cov( X , Y )

cov( X , Y ) =0   Z2  (1  9) X2  10 X2  29.16667

7)
y y y y

x x x x
0 1
(a) (b) (d)


a) f X x    f ( x, y) dy
 X ,Y
 For x<0 or x > 1 : f X x  = 0

For 0< x < 1:

2
2 2 1 1 2
f X x    ( x  2 y ) dy  ( )( )( x  2 y ) 2
1
7 7 2 2 y 1

1 1 1 2

14
 14

( x  4) 2  ( x  2) 2  (8 x  16)  ( 4 x  4)  ( 4 x  12)  ( 2 x  3)
14 7

Therefore

2
 ( x  3) 0  x 1
f X x    7
0 otherwise

b) fY  y    f ( x, y ) dx
 X ,Y

For y<1 or y > 2 : fY  y  = 0

For 1< y < 2:

1
2 2 1 1 1 1
fY  y    ( x  2 y ) dx  ( )( x  2 y ) 2
7 7 2 x0 7
 
 (1  2 y ) 2  ( 2 y ) 2  (1  4 y )
7
0

Therefore

1
 (4 y  1) 1 y  2
fY  y    7
0 otherwise

 1
2 2 2 x3 3 2 1 2 1 3 229
c)  X  E X   xf
 X ( x ) dx  0 7 ( x  3 x ) dx  (  x )  (  )  
7 3 2 0 7 3 2 7 6 

11
  0.523809523
21
 2
1 1 4 3 y 2 2 1  32 4 1 
Y  EY   yf
 Y ( y ) dy  1 7 ( 4 y 2
 y ) dy  ( y  )  (  2 )  (  ) 
7 3 2 1 7 3 3 2 

1  38 11  1  65  65
  )     1.547619
7  3 6  7  6  42
   2 1
2
d) E XY     xy f ( x, y )dxdy    y ( x 2  2 xy) dxdy
   X ,Y y 1 x0
7

2 2 2
2 x3 2 1 2 1 2 y 2
 y 1 7 y ( 3  x y) x  0dy  1 7 y ( 3  y)dy   7 (3  y
y 1
)dy

2 y 2 y3 2 2  2 8 1 1  2 1  2  17  17
 (  )  (  )  (  )    3        0.809524
7 6 3 1 7 3 3 6 3  7 6  7  6  21

17  11  65  1
cov( X , Y )  E XY    X  y       0.00113379  
21  21  42  882

e) cov( X , Y )  0  X and Y are not uncorrelated  X and Y are not

statistically independent.
  2 1
x  2
f) Eg ( XY )    g ( xy) f ( x, y )dxdy    (y 3
 x 2 y )   ( x  2 y) dxdy
  X ,Y y 1 x  0 7
2 1 2
2  3 x2 x  2  x4 x3 2 x2 1
  x y   2x 2 y 2  2 2  dxdy    y  3  x3 y 2  2  dy
7 y1 x 0  y 3
y  7 y 1  4 3y 3 y x  0
2
2 1 1 2 1  21 1 2 12
   y  3  y 2  2 dy   y 2  2  y3  
7 y 1  4 3y 3 y  7 8 6y 9 yy 1

2  1 1 16 1   1 1 2  2   3  128  9  12  16  72 
           1   
7  2 24 9 2   8 6 9  7  72 

2  184  46
    0.730159
7  72  63