Solutions for Sheet-2 part 2

1] ⎧⎪ce−2x x ≥0
f (x) = ⎨
⎪⎩0 x <0

Find c. What is P(X > 2)?

Solution:
S l ti f( ) is
f(x) i the
th density
d it function
f ti off an exponential
ti l random
d variable
i bl
with parameter λ = 2 , and

F(x ) = 1 − e− λx = 1 − e − 2 x x≥0

Th , P(X > 2 ) = 1-
Then 1 (1 -44) =e-44
1 F(2) = 1-(1-e
Another solution: ∞
((a)) Since,, f(x)
( ) is a pp.d.f.,, we must have that ∫ f (x )dx = 1
−∞
∞
Implying that c ∫ e − 2 x dx = 1
0

− ∞
Or, −c e 2 x =1
2 x =0

O c = 2 , andd then
Or, h f (x) = 2e−2x x ≥0
∞ ∞
−2x −2x ∞
(b) P ( X > 2 ) = ∫ f ( x ) dx = ∫ 2 e dx = − e = e−4
x=2
2 2
2]
⎧⎪a + bx 2 0 ≤ x ≤1
f (x) = ⎨
⎪⎩ 0 otherwise
If E [ X ] = 3
If,
5
Find ((i)) a,, b ((ii)) Var(X).
( )
Solution: ∞
(i) Since, f(x) is a p.d.f., we must have that ∫ f ( x )dx = 1
1 −∞ 1
2 b
Implying that ∫ ( a + bx ) dx = 1 Then, ax + x 3 ⎤
⎥ = 1
0 3 ⎦0
Then,
b
a + = 1 (1)
3
Also we have
Also,
 1 1 1
2 a 2 b 4⎤ 3
E(X) = ∫ xff (x)dx
d = ∫ x(a +bx
b )dxd = x + x ⎥ =
2 4 ⎦0 5
0 0

Then,
a b 3 (2)
+ =
2 4 5

From equations (1) and (2) , we have

3 6
a= , b=
5 5
(ii) 1 1 1
2 2 4
3 6 x 3 x 6 x
E ( X 2 ) = ∫ x 2 f ( x ) dx = ∫ x 2 ( + ) dx = ∫ ( + ) dx
0 0
5 5 0
5 5
1
1 3 6 5⎤ 11
= x + x ⎥ =
5 25 ⎦ 0 25
Then,
11 9 2
Var ( X ) = E [ X ] − [E ( X ) ]
2 2
= − =
25 25 25
3] ⎧ 20000
⎪ x3 x > 100
⎪
f (x) = ⎨
⎪ 0 elsewhere
l h .
⎪
⎩
∞ ∞ ∞
20000 −2 ∞ 10000
(a) P ( X > 2000 ) = ∫ f ( x ) dx = ∫ dx = − 10000 x = −
2000 2000 x3 2000 x2 2000

⎛ 10000 ⎞ 1
= − ⎜⎜ 0 − ⎟
2 ⎟
= = 0 .0025
⎝ ( 2000 ) ⎠ 400

300 300 300

20000 − 2 300 10000
(b) P (150 < X < 300 ) = ∫ f ( x ) dx = ∫ x3
dx = − 10000 x
150
= −
x2
150 150 150

⎛ 10000 10000 ⎞ ⎛ 100 1 ⎞ 675

= − ⎜⎜ 2
− ⎟⎟ = ⎜ − ⎟= − 0 .33
⎝ ( 300 ) (150 ) 2 ⎠ ⎝ 225 9 ⎠ 2025
(c)
∞ ∞ ∞
20000 −1 ∞ 20000
(i) E ( X ) = ∫ xf ( x ) dx = ∫ 2
dx = − 20000 x =−
100 100 x 100 x 100

⎛ 20000 ⎞
= −⎜ 0 − ⎟ = 200
⎝ 100 ⎠
Then, E(3X-1) = 3E(X) – 1 = 3(200)-1 = 599

∞ ∞
(ii) E ( X 2 ) = 2 20000 ∞
∫x f ( x ) dx = ∫ x
dx = 20000 ln x 100 = ∞
100 100

Then E ( X 2 ) does not exist and then Var(X) does not exist

Then , Var(-2X + 5) does not exist

4] The time it takes to repair a personal computer is a random variable
whose density, in hours, is given by
⎧⎪ 1 0<x<2
f (x) = ⎨ 2
⎪⎩ 0 otherwise
The cost of the repair depends on the time it takes and is equal to
40 + 30 x when the time is x. Computep the expected
p cost to repair
p a
personal computer.

Solution :
The expected cost to repair a personal computer
= E[40 + 30 x ] = 40 +30
30 E[ x ] (1)
Now, 2
E[ X]= ∫ x f ( x ) dx
0

() (12 )dx
2 2 1
= ∫ x 1 dx = ∫ x 2
2
0 0
3 3 2
2
2 x dx = x
2 2 8 = 0 . 943 (2)
= ∫ 3 2 3
=
3
0 0
From (1), (2) we have

= E[ 40 + 30 x ] = 40 +30 E[ x ] = 40 +(30)(0.943)=68.28
5] If X is a continuous random variable having distribution
function F , then its median is defined as that value of m for
which
F(m)
( )=½
Find the median of the random variables with density function
−x
(a) f ( x ) = e x≥0

(b) f(x) = 1 , 0 ≤ x ≤ 1 .
Solution :
( )
(a) m m
F ( m ) = P ( X ≤ m) = ∫ f ( x ) d x = ∫ e − x dx
−∞ 0
− x m − m
= −e =1− e
0

−m 1 so, e−m = 1
Now, F(m) = ½ , then 1− e = 2
2

Then ,
m = −ln((1) = ln2
2
(b)

m m
F ( m ) = P ( X ≤ m) = ∫ f ( x ) d x = ∫ (1)dx
−∞ 0
m
=x0 =m

Now, F(m) = m = ½ , then

m=½
6] ⎧ ( z − 8) for 8 ≤ z ≤ 9
⎪
f ( z ) = ⎨(10 − z ) for 9 < z ≤ 10
⎪ 0 otherwise
⎩

(a) Find the cumulative distribution function F(x)

((b)) Calculate the mean and variance of the random variable X.
Solution :
(a)
i) For 8 ≤ z ≤ 9
2 z
z z
t
F ( z ) = ∫ f ( t ) dt = ∫ ( t − 8 ) dt = ( − 8t)
−∞ 8 2 8

z2 z2
= ( − 8 z ) − ( 32 − 64 ) = ( − 8 z ) + 32
2 2
z 2 − 16 z + 64 (z − 8)2
= = , 8 ≤ z ≤ 9
2 2
ii) For 9 ≤ z ≤ 10 2 z
z 1z 1 (10 − t )
F ( z ) = F ( 9 ) + ∫ f ( t ) dt = + ∫ (10 − t ) dt = +
9 2 9 2 −2 9
1 1
[ 2 1 1
] 2
= − (10 − z ) − 1 = − (10 − z ) +
2 2 2 2
[
1
2
]
(10 − z ) 2
=1− 9 ≤ z ≤ 10
2
Then,

⎧ 0 z<8
⎪ (z − 8) 2
⎪⎪ 8≤z<9
F(z) = ⎨ 2
2
⎪1 − (10 − z )
9 ≤ z < 10
⎪ 2
⎪⎩ 1 z ≥ 10
 ∞ 10
(b) E ( X ) =
∫ z f ( z ) dz = ∫ z f ( z ) dz
−∞ 8
9 10
= ∫ z ( z − 8 ) dz + ∫ z (10 − z ) dz
8 9
9 9
2 2
= ∫ ( z − 8 z ) dz + ∫ (10 z − z ) dz
8 8
9 10
z3 z 3
=( − 4z2 ) + (5 z 2 − )
3 3
z =8 z=9
 [ ][
= ( 243 − 324 ) − ( 512 − 256 ) + (500 − 1000 ) − ( 405 − 243 )
3 3
]
= 13 + 14 = 27 = 9
3 3 3

Then
h , E[X]
[ ]=9 (1)
 ∞ 10
E (X 2 ) = ∫ z 2 f ( z ) dz =
∫ z 2 f ( z ) dz
−∞ 8
9 10
= ∫ z 2 ( z − 8 ) dz + ∫ z 2 (10 − z ) dz
8 9
9 9
= ∫ ( z 3 − 8 z 2 ) dzd + ∫ (10 z 2 − z 3 ) dz d
8 8
4 9 4 10
z 8 3 10 3 z
=( − z ) +( z − )
4 3 3 4
z =8 z=9
 [ 4 3
][
= ( 6561 − 1944 ) − (1024 − 4096 ) + (10000 − 2500 ) − ( 2430 − 6561 )
3 4
]
= 451 + 523 = 974 = 487 = 81 .17
12 12 12 6

Then [ 2] = 81.17
h , E[X 81 1 (2)

From , (1) , (2) we have

487 1
Var ( X ) = E[ X ] − [E ( X ) ] =
2 2 − 81 = = 0 .17
6 6
7] Suppose that X is uniformly distributed over [ 0 , 2] ,
evaluate the followingg
(a) P(X < ½ )
(b) P(½ < X < 1 )
(c) P(X > 0.75 )
(d) E[X] , Var[X]

Solution :
⎧⎪ 1 0<x<2
f (x) = ⎨ 2
⎪⎩ 0 otherwise

The cumulative distribution function is given by

 ⎧0 x<0
⎪x
F(x ) = ⎨ 0≤x<2
⎪2
⎩1 x≥2

(a) P(X < ½ )= F(½) = ¼

(b) P(½ < X < 1 ) = F(1) – F(½) = ½ - ¼ = ¼

0 . 75 5
( ) P(X > 0.75
(c) 0 75 ) = 1-
1 F(0.75)
F(0 75) = 1− =
2 8

α +β 2
(d) E(X) = μ = 2
= 2
= 1
 2 2
2 (β−α) ( 2) 4 1
Var(X) = σ = 12
= 12
= 12 = 3
Another solution :
1 1
2 2
1 1 1 1 . 12 = 1
( )
(a) P (X < 2
) = ∫ 2
dx = 2 ∫ dx = 2 4
0 0
1
(b) P ( 12 < X < 1 ) = ∫ 1
2
dx = 1
2
. 12 = 1
4
1
2

2
(c) P ( X > 0 . 75 ) = 1 1 1.5 5
∫ 2
dx = 2
( 2 − 0 . 75 ) = 2 4
= 8
0 . 75
8] Suppose that X is uniformly distributed over [ 0 ,α] ,
determine α so that the followingg are satisfied
(a) P(X >1) = 13

(b) P(X < 12) = 0.7

Solution :
⎧1 0<x<α
f (x) = ⎨ α
⎩0 otherwise

The cumulative distribution function is given by

 ⎧0 x<0
⎪x
F(x ) = ⎨ 0≤x<α
⎪α
⎩1 x≥α

(a) P ( X > 1 ) = 1 − F (1 ) = 1 − 1 = 1
α 3

1 2 α = 3
th
then,
α
= 3
then
then,
2

(b) P (X < 1 ) = F( ) = 1 1 = 0 .7
2 2 2α

then
then, α= 0 .7 = 7 = 0 .35
2 20
9] If X is a normal random variable with parameters µ = 10 ,
σ2 = 36 , compute
p
(a) P{ X > 5 } ;
(b) P{ 4 < X < 16 } ;
(c) P{ X < 8 } ;
(d) P{ X < 20 } ;
(e) P{ X > 16 } .
Solution : µ = 10 , σ2 = 36 , so σ = 6 .
( ) P(X > 5) = P( X −μ > 5−10 ) = P( Z > −0.83)
(a)
σ 6
= 1 − Φ(−0.83) = Φ(0.83) = 0.7967

4 −10 X −μ 16−10
((b)) P(4 < X < 16) = P( < < ) = P(−1 < Z < 1)
6 σ 6
= Φ(1) − Φ(−1) = Φ(1) − [1 − Φ(1)]
= 2Φ(1) − 1 = 2(0.8413) − 1 = 0.6826

(c) P(X < 8) = P( X −μ < 8−10 ) = P( Z < −0.33)

σ 6
= Φ(−0.33) = 1 − Φ(0.33) = 1 − 0.6293 = 0.3707
 X − μ 20 −10
((d)) P ( X < 20 ) = P ( < ) = P ( Z < 1 . 67 )
σ 6
= Φ (1 . 67 ) = 0 . 9525

((e)) P ( X > 16 ) = P ( X − μ > 16 −10 ) = P ( Z > 1)

σ 6
= 1 − Φ (1) = 1 − 0 . 8413 = 0 . 1587
10] A lawyer commutes daily from his suburban home to his
midtown office. The average time for a one one-way
way trip is 24
minutes, with a standard deviation of 3.8 minutes .
Assume
ssu e tthee ddistribution
st but o ofo trip
t p times
t es to be normally
o a y
distributed.

(a) What is the probability that a trip will take at least ½ hour?

(b) If the office opens at 9:00 A.M., and he leaves his house at
8:45 A.M. daily,y, what ppercent of the time is he late for work?

( ) If he leaves the house at 8:35 A.M. and coffee is served at

(c)
the office from 8:50 A.M., until 9:00 A.M., what is the
probabilityy that he misses coffee ?
p
(d) Find the length of time above which we find 15% of the
trips.
trips

(e) Find the probability that 2 of the next 3 trips will take at
least ½ hour.

Solution:
μ = 24,
24 σ = 3.8
38
(a) P [X >30] = P [Z > 1.58]
1 φ(1 58)
= 1-φ(1.58)
=1- 0.9429
= 0.0571
0 0571
(b) P [X >15] = P [Z > -2.37] = 1-φ(-2.37) =φ(2.37) = 0.9911

(c) P( He misses coffee ) = 1-P (15 ≤ X ≤ 25 ) (1)

Now, we have to find P (15 ≤ X ≤ 25 )

X −μ
P(15 < X < 25) = P(153−.824 < σ
< 253−.824 ) = P(−2.37 < Z < 0.26)
= Φ(0.26) − Φ(−2.37) = Φ(0.26) − [1 − Φ(2.37)]
= Φ(0.26) + Φ(2.37) − 1
= 0.6026 + 0.9911 − 1
= 0.5937 (2)
From (1), (2) we get
P( He misses coffee ) = 1-0 .5937=0.4063
(d) P(X > k) = 0.15, then P(X < k) = 0.85
X −μ
P (X < k ) = P ( σ < k 3−. 24 8
) = P(Z < k − 24
3 .8
)
= Φ ( k 3−. 24
8
) = 0 . 85

Then
Then, k − 24 = 1 . 04 and so k =27.95
27.95
3 .8
((e)) It iis bi
binomial
i l with
ith parameters
t n = 3 , andd
p = P(X ≥ 30 ) = 0.0571 (from part (a)) , then

⎛3⎞ 2 1
P [ X = 2 ] = ⎜⎜ ⎟⎟ ( 0 . 0571 ) ( 0 . 9429 ) = 0 . 00922
⎝2⎠
11] The Scholastic Aptitude Test mathematics test scores
across the population of high school seniors follow a normal
distribution with mean 500 and standard deviation 100 . If five
seniors are randomly chosen , find the probability that
(a) All scored below 600
(b) Exactly three of them scored above 640 .
Solution : µ = 500 , σ = 100 .
( ) First
(a) Fi t , we have
h to
t find
fi d
X − μ 600 − 500
P ( X < 600 ) = P ( < ) = P ( Z < 1)
σ 100
= Φ (1) = 0 . 8413
The number of seniors with scored below 600 is distributed as
binomial with parameters n = 5 , p = 0.8413, so the probability
that all the five seniors scored below 600 is given by

⎛5⎞ 5 0
P[ X = 5] = ⎜⎜ ⎟⎟ (0.8413) (0.1587 ) = 0.4214
⎝5⎠
(b) First , we have to find
X − μ 640 − 500
P ( X > 640 ) = P ( > ) = P ( Z > 1 .4 )
σ 100
= 1 − Φ (1 . 4 ) = 1 − 0 . 9192 = 0 . 0808

The number of seniors with scored above 640 is distributed as

binomial with pparameters n = 5 , p = 0.0808,, so the probability
p y
that exactly three of them scored above 640 is

⎛5⎞ 3 2
P[ X = 3] = ⎜⎜ ⎟⎟ (0.0808 ) (0.9192 ) = 0.0045
⎝ 3⎠
12] The annual rainfall ( in inches) in a certain region is normally
distributed with µ = 40 ,,σ = 4 . What is the pprobability
y that in 2 of the
next 4 years the rainfall will exceed 50 inches ? Assume that the rainfalls
in different years are independent .

Solution : µ = 40 , σ = 4 .
First, we have to find P(X > 50) as follows

X −μ
P ( X > 50 ) = P ( > 50 − 40 ) = P ( Z > 2 . 5 )
σ 4
= 1 − Φ ( 2 . 5 ) = 1 − 0 . 9938 = 0 . 0062
The number of years with rainfall exceed 50 inches is binomial with
parameters n = 4 and p = 0.0062
0 0062 , so

⎛4⎞
P [ X = 2 ] = ⎜⎜ ⎟⎟ ( 0 . 0062 ) 2 ( 0 . 9938 ) 2 = 0 . 00023
⎝2⎠
13]A certain type of light bulb has an output that is normally distributed
with mean 2,000 end foot candles and standard deviation 85 end foot
candles . Determine a lower specification limit L so that only 5 percent
of the light bulbs produced will be defective . ( That is , determine L so
that P{ X ≥ L } = 0.95
0 95 , where X is the output of a bulb .))
Solution : µ = 2000 , σ = 85 .

X −μ
P(X ≥ L) = P( σ ≥ L − 2000
85 ) = 1 − Φ ( L −85
2000
) = 0.95
Then.
Φ ( L − 2000 ) = 0 . 05
85
and , we have Φ(-1.645)=0.05 , then

L − 2000 = − 1 . 465
46
85

and , L = 2000-(1.645)(85)= 1860.175 ≈ 1866

Another Solution :
Since L − 2000
Φ( ) = 0 . 05 < 0 . 5
85
Then,,
Φ ( − L − 2000 ) = 1 − 0 . 05 = 0 . 95
85
i
i.e.,
Φ ( 2000 − L ) = 0 . 95
85
and , we have Φ(1.645)=0.95 , then
2000 − L = 1 . 645
85

and , L = 2000
2000-(1
(1.645)(85)
645)(85)= 1860.175
1860 175 ≈ 1866
14] A manufacturer produces bolts that are specified to be between 1.19
and 1.21 inches in diameter . If its production process results in a bolt’s
diameter being normally distributed with mean 1.20 inches and standard
deviation 0.005 , what percentage of bolts will not meet specifications ?
Solution : µ = 1.2 , σ = 0.005
First we have to find the probability that the bolts will meet
specifications , which equals

1 . 19 − 1 . 20 X − μ 1 . 21 − 1 . 20
P (1 . 19 < X < 1 . 21 ) = P ( < < )
0 . 005 σ 0 . 005
= P (−2 < Z < 2) = Φ (2) − Φ (−2)
= Φ ( 2 ) − [1 − Φ ( 2 )] = 2 Φ ( 2 ) − 1
= 2 ( 0 . 9772 ) − 1 = 0 . 9544

Then the probability that the bolts will not meet specifications equals
1 0.9544
1- 0 9544 = 0.0456
0 0456 so that its percentage equals 4.56
4 56 %
15] Suppose the amount of time a light bulb works before burning out
is a normal random variable with mean 400 hours and standard
deviation 40 hours. If an individual purchases two such bulbs, one of
which will be used as a spare to replace the other when it burns out,
what is the probability that the total life of the bulbs will exceed 750
hours?
Solution : X1, X2 are the life time of each bulb & X = total life of the
bulbs . Now, we have to find the mean and the standard deviation of X,
.e., μX aandd σX
i.e.,
E(X) = μ X = μ X1 + μ X 2 = μ + μ = 2μ = 800

Var (X) = σ2X = Var (X1 ) + Var(X 2 ) = σ2 + σ 2 = 2σ 2 = (2)(40) 2 = 3200

Then we have σ X = Var ( X ) = 3200 = 56 .6
Then,

X − μ X 750 − 800
P(X > 750) = P( > ) = P( Z > −0.88) = 1 − Φ(−0.88) = Φ(0.88) = 0.8106
σX 56.6
16] The time (in hours ) required to repair a machine is an exponentially
distributed random variable with parameter
λ=2.
(a) What is the probability that a repair time exceeds 5 hours ?
(b) What is the conditional probability that a repair takes at
least 9 hours , given that its duration exceed 5 hours ?
Solution : λ = 2 ,
Note that , for an exponential random variable, we have
f ( x ) = λ e − λx = 2e − 2 x x≥0
F( x ) = 1 − e − λ x = 1 − e − 2 x x≥0
((a)) P ( X > 5 ) = 1 − F ( 5 ) = 1 − (1 − e − 10 ) = e − 10
(b) By using the memoryless property

P(X > 9 X > 5) = P(X > 4 + 5 X > 5)

= P(X > 4) = 1 − F(4) = 1 − (1 − e−8 ) = e−8
17]The number of years a radio functions is exponentially
distributed with pparameter λ = 18 . If Jones buys
y a used radio ,
what is the probability that it will be working after an
additional 10 yyears ?

Solution : λ = 1
8
Note that
− λ x − ( 1 )x
f ( x ) = λe = 1e 8 x≥0
8
−
F( x ) = 1 − e−λx = 1 − e 8
( 1 )x
x≥0
The, required probability is
−10 ( 18 ) −1 . 25
P ( X > 10 ) = 1 − F(10 ) = 1 − (1 − e )=e