Statistics for Data Science - 2

Week 3 Graded assignment solutions

1. Suppose 1 in 100 products that are coming out of a production line is defective. Sup-
pose we randomly pick and keep aside products from the production line till the first
defective item is obtained. Let the random variable X represent the number of prod-
ucts that are kept aside (Assume that the first defective item is also kept aside). Find
Var(X).
1
(a)
100
99
(b)
100
(c) 100
(d) 9900

Solution:
The random variable X represent the number of products that are kept aside (including
the first defective item) before the first defective is obtained.
It is given that 1 out of 100products
 are defective.
1
Therefore, X ∼ Geometric
100
Now,
1−p
Var(X) =
p2
1− 1
= 1 100 = 9900
( 100 )2

Hence, the correct option is (d).

2. Two coins are tossed. The probabilities of occurrence of tail on the first and the second
coin are 0.6 and 0.4, respectively. If the random variable X represents the number of
heads obtained, find the expected value of X. (Enter the answer correct to 2 decimal
points).
Answer: 1
Solution:

1
 Given,

P (tail occurs on the first coin) = 0.6. (1)

P (tail occurs on the second coin) = 0.4. (2)

Random variable X denote the number of heads obtained after the tossing of two coins.
Therefore, X will take the values in {0, 1, 2}.
Now,
X
E(X) = xP (X = x)
x∈X
=0.P (X = 0) + 1.P (X = 1) + 2.P (X = 2)

P (X = 1) =P (H on first coin and T on second coin) + P (T on first coin and H on second coin)
=(0.4 × 0.4) + (0.6 × 0.6)
=0.52

P (X = 2) =P (H on both the coins)

= (0.4 × 0.6) = 0.24

Therefore, E(X) = 0.52 + (2 × 0.24) = 1

3. Let the two random variables X and Y be independent with means equal to 10 and
20, and variances equal to 2 and 4, respectively. Find the value of Var(XY ).
Hint: If X and Y are independent, X 2 and Y 2 are also independent.
Answer: 1208
Solution:
Mean and variance of X is 10 and 2, respectively.
Mean and variance of Y is 20 and 4, respectively.

Var(XY) =E[(XY )2 ] − (E[XY ])2

=E[X 2 Y 2 ] − (E[X]E[Y ])2 , X and Y are independent.
=E[X 2 ]E[Y 2 ] − E[X]2 E[Y ]2 , X and Y are independent.
=(V ar(X) + E[X]2 )(V ar(Y ) + E[Y ]2 ) − E[X]2 E[Y ]2
=(2 + 102 )(4 + 202 ) − 102 202
=(102 × 404) − 40000
=41208 − 40000
=1208

2
4. Let X and Y be two independent discrete random variables. Define random variables
U and V as

X − E(X) Y − E(Y )
U= , V =
SD(X) SD(Y )
Find Cov(U, V ).
Answer: 0
Solution:
Cov(U, V ) = E(U V ) − E(U )E(V ).
Since U and V are the standardized form of random variables X and Y , respectively,

E(U ) = E(V ) = 0 and V ar(X) = V ar(Y ) = 1

Now,

Cov(U, V ) = E(U V )
  
X − E(X) Y − E(Y )
=E
SD(X) SD(Y )
1
= E[(X − E(X))(Y − E(Y ))]
SD(X)SD(Y )
= E[XY − XE(Y ) − Y E(X) + E(X)E(Y )]
= E[XY ] − E[X]E(Y ) − E[Y ]E(X) + E(X)E(Y )

Since X and Y are independent, E(XY ) = E(X)E(Y )

Therefore, Cov(U, V ) = E[X]E(Y ) − E[X]E(Y ) = 0.
Use the following information to answer questions (5) and (6).
Number of people (X) who make a reservation in a restaurant a day is a random
variable with mean equal to 10 and variance equal to 2.

5. Using Markov’s inequality, find a bound on the probability that on a particular day,
the number of reservations will exceed 30.
1
(a) P (X > 30) ≤
4
1
(b) P (X > 30) ≥
3
10
(c) P (X > 30) ≤
31
10
(d) P (X > 30) >
31
Solution:

3
 Random variable X represents the number of people who make reservation in a restau-
rant. It is given that
E(X) = 10 (3)
Using Markov’s inequality, we know that
µ
P (X ≥ c) ≤
c
10
Therefore, P (X > 30) = P (X ≥ 31) ≤ .
31
Therefore, the correct option is (c).
6. Find a bound on the probability that on a particular day, number of reservations made
will lie in between 6 and 14 using Chebyshev’s inequality.
7
(a) P (6 < X < 14) ≤
8
7
(b) P (6 < X < 14) ≥
8
7
(c) P (6 < X < 14) >
8
1
(d) P (6 < X < 14) ≤
8
Solution:
Using the Chebyshev’s inequality, we know that
1
P (| X − µ |≥ kσ) ≤ (4)
k2
1
P (µ − kσ < X < µ + kσ) ≥ 1 − (5)
k2
Given µ = 10 and σ 2 = 2
Now, we can write P (6 < X < 14) as
1
P (10 − kσ < X < 10 + kσ) ≥ 1 − 2 . Using (5)
k
Now, let
10 − kσ =6 (6)
10 + kσ =14 (7)
Solving (6) and (7), we get kσ = 4

4 16
⇒k= ⇒ k2 = =8
σ 2
1 7
Therefore, P (6 < X < 14) ≥ 1 − =
8 8
Hence, the correct option is (b).

4
7. The joint probability mass function of three discrete random variables X, Y and Z is
given as
1
p(0, 1, 2) = p(0, 2, 3) = p(1, 0, −2) =
3
Calculate Var(XY + 2Z).
52
(a)
9
32
(b)
9
80
(c)
3
56
(d)
3
Solution:
t1 t2 t3 t1 t2 + 2t3 fXY Z (t1 , t2 , t3 )
0 1 2 4 1/3
0 2 3 6 1/3
1 0 −2 -4 1/3

Joint PMF of X, Y and Z.

1
XY + 2Z will take the values in {-4, 6, 4} with the probabilities each.
3
1
E(XY + 2Z) = [−4 + 6 + 4]
3
6
= =2
3
1
E[(XY + 2Z)2 ] = [(−4)2 + 62 + 42 ]
3
1
= [16 + 36 + 16]
3
68
=
3
Now,
Var(XY + 2Z) =E[(XY + 2Z)2 ] − [E(XY + 2Z)]2
68
= − 22
3
56
=
3
Hence, the correct option is (d).

5
8. An urn contains 5 white balls and 5 red balls. 2 balls are selected at random. Let
X denote the number of red balls drawn and let Y denote the number of white balls
drawn. Find the correlation coefficient between X and Y .
(a) ρ(X, Y ) = 1
(b) ρ(X, Y ) = −1
(c) ρ(X, Y ) = 0
(d) ρ(X, Y ) = −0.5
Solution:
Two balls are selected at random from the urn containing 5 white and 5 red balls.
Random variable X represent the number of red balls drawn.
Therefore, X will take values in {0, 1, 2}.
Random variable Y represent the number of white balls drawn.
Therefore, Y will take values in {0, 1, 2}.
Joint probability distribution of X and Y is given by

X
0 1 2
Y
10
0 0 0
45
25
1 0 0
45
10
2 0 0
45

Joint distribution of X and Y .

Now,
     
10 25 10
E(X) = 0 × + 1× + 2×
45 45 45
=1

Similarly, E(Y ) = 1.

     
2 10 25 2 10
E(X ) = 0 × + 1× + 2 ×
45 45 45
65
=
45

6
 65
Similarly, E(Y 2 ) =.
45
65 20
Now, Var(X) = Var(Y ) = − (1)2 =
45 45
   
10 25
E(XY ) = 0 × + 1× + (2 × 0)
45 45
25
=
45

Correlation coefficient between X and Y is given by

Cov(X, Y )
ρ(X, Y ) =
SD(X)SD(Y )
E(XY ) − E(X)E(Y )
= p
Var(X)Var(Y )
( 25 − 1)
= q 45
( 20
45
) × ( 20
45
)
=−1

Therefore, the correct option is (b).

9. Five students each from class 8, 9 and 10 have been nominated for the formation of
the school committee. The number of boys and girls who are selected from each of the
classes is given in Table 4.1.A.

Class 8 Class 9 Class 10

Girls 2 2 3
Boys 3 3 2

Table 4.1.A: Total number of boys and girls selected.

If the committee comprises of two students from each class, find the expected number
of girls in the committee. (Enter the answer correct to 1 decimal point)
Answer: 2.8
Solution:
Let X1 represent the number of girls from class eight in the school committee.
Let X2 represent the number of girls from class nine in the school committee.
Let X3 represent the number of girls from class ten in the school committee.
We need to find E(X1 + X2 + X3 ).

7
We know that E(X1 + X2 + X3 ) = E(X1 ) + E(X2 ) + E(X3 ).

Since total number of girls selected from class eight is 2, therefore, the committee can
comprise of either 0 girl or 1 girl or 2 girls from class eight.
i.e. X1 will take values in {0, 1, 2}.
Now
3
C2 3
P (X1 = 0) = 5 =
C2 10
3 2
C1 × C1 6
P (X1 = 1) = 5 =
C2 10
2
C2 1
P (X1 = 2) = 5 =
C2 10

     
3 6 1 8
Therefore, E(X1 ) = 0× + 1× + 2× =
10 10 10 10

Similarly, total number of girls selected from class nine is 2, therefore, the committee
can comprise of either 0 girl or 1 girl or 2 girls from class nine.
8
i.e. X2 will take values in {0, 1, 2}, hence E(X2 ) = .
10

Total number of girls selected from class ten is 3 and we have to select 2 students from
each class, therefore, the committee can comprise of either 0 girl or 1 girl or 2 girls
from class ten.
i.e. X3 will take values in {0, 1, 2}.
Now
2
C2 1
P (X3 = 0) = 5 =
C2 10
3 2
C1 × C1 6
P (X3 = 1) = 5 =
C2 10
3
C2 3
P (X3 = 2) = 5 =
C2 10

     
1 6 3 12
Therefore, E(X3 ) = 0× + 1× + 2× =
10 10 10 10

8
 Now

E(X1 + X2 + X3 ) =E(X1 ) + E(X2 ) + E(X3 )

8 8 12
= + +
10 10 10
=2.8

Hence, expected number of girls in the class committee is 2.8.

10. A share of a company costs |1000 today. Suppose today’s share price increases by
50% with probability 0.6 and decreases by 50% with probability 0.4. Independent of
today, suppose that tomorrow’s share price increases by 20% with probability 0.2, and
decreases by 30% with probability 0.8. If you decide to buy 3 shares today, find the
expected profit (in |) at the end of 2 days.

(a) −120
(b) 360
(c) 120
(d) −360

Solution:
The cost price of a share of the company is |1000.
Let the random variable X represent the price of the share at the end of 2 days.
Price can either go up by 50% with probability 0.6 or can go down by 50% with prob-
ability 0.4 on the first day.
Independent of today, the share price can either go up by 20% with probability 0.2 or
can go down by 30% with probability 0.8.

i.e. If the share price increases by 50% on the first day, the price of the share will
become |1500.
Andthe price of the share at the end of two days if the share prices increases by 20%
20
is | 1500 × + 1500 =|1800 with probability (0.6 × 0.2) = 0.12.
100
Similarly,the price of the share
 at the end of two days if the share prices decreases by
30
30% is | 1500 − 1500 × =|1050 with probability (0.6 × 0.8) = 0.48.
100

Again, if the share price decreases by 50% on the first day, the price of the share will
become |500.
Andthe price of the share
 at the end of two days if the share prices increases by 20%
20
is | 500 × + 500 =|600 with probability (0.4 × 0.2) = 0.08.
100
Similarly, the price of the share at the end of two days if the share prices decreases by

9
  
30
30% is | 500 − 500 × =|350 with probability (0.4 × 0.8) = 0.32.
100

Therefore, X will take values in {1800, 1050, 600, 350}, where

P (X = 1800) = 0.12
P (X = 1050) = 0.48
P (X = 600) = 0.08
P (X = 350) = 0.32
Now,

E(X) =(1800 × 0.12) + (1050 × 0.48) + (600 × 0.08) + (350 × 0.32)

=880

The expected gain at the end of two days if you buy one share is |(880-1000) = -|120.
Therefore, if you buy 3 shares of the company, expected gain will be -|360.

Hence, the correct option is (d).

11. A lottery has 500 tickets out of which only 2 tickets contain prizes worth |500 and
|1,000; the rest are worth |0. If one has bought 2 tickets, what will be his/her ex-
pected gain (in |)?
Answer: 6

Solution:
In the lottery, only two tickets out of 500 contain prizes worth |500 and |1,000.
If one has bought two tickets, one can get the prizes worth |0, |500, |1,000 and |1,500.
Let the random variable X represent the worth of the prizes of two tickets.
Therefore, X will take values in {0, 500, 1000, 1500}.

498
C2
P (X = 0) = P (Both the tickets are worth |0) = 500
C2
498
C 1 1C 1
P (X = 500) = P (One of the ticket is worth |0 and the other is worth |500) = 500
C2
498
C 1 1C 1
P (X = 1000) = P (One of the ticket is worth |0 and the other is worth |1000) = 500
C2
P (X = 1500) = P (One of the ticket is worth |500 and the other is worth |1000) =
2
C2
500
C2

10
 498 498
C 1 1C 1 498
C 1 1C 1 2
       
C2 C2
E(X) = 0 × 500 + 500 × 500 + 1000 × 500 + 1500 × 500
C2 C2 C2 C2
1
= 500 [500 × 498C 1 + 1000 × 498C 1 + 1 × 1500]
C2
1
= 500 [249000 + 498000 + 1500]
C2
748500
= =6
124750
Therefore, the expected gain is |6.

12. If X and Y are independent random variables with E[X] = 1, Var(X) = 2, E[Y ] = 3,
and Var(Y ) = 4, then find E[(2X + 3Y )2 ].
Answer: 165
Solution:
Mean and variance of X is 1 and 2, respectively.
Mean and variance of Y is 3 and 4, respectively.

E[(2X + 3Y )2 ] = E[4X 2 + 9Y 2 + 12XY ]

= 4E[X 2 ] + 9E[Y 2 ] + 12E[XY ]
= 4{V ar(X) + (E[X])2 } + 9{V ar(Y ) + (E[Y ])2 } + 12E[X]E[Y ]
X and Y are independent.
E[(2X + 3Y )2 ] = 4{2 + (1)2 } + 9{4 + (3)2 } + 12(1)(3)
= 4{2 + 1} + 9{4 + 9} + 36
= 4 × 3 + 9 × 13 + 36
= 165

13. Suppose X and Y are random variables such that X follows a discrete uniform distri-
bution on the set {0, 1} and Y = X 2 . Determine Cov(X, Y ). Enter the answer correct
to two decimal places.
Answer: 0.25
Solution:
For finding the Cov(X, Y ), firstly we have to find E[X], E[Y ] and E[XY ].
X 1 1 1
E[X] = x · P (X = x) = 0 · +1· =
x
2 2 2

11
 X Y = X2 P (Y = y)
1
0 0 P (Y = 0) = P (X 2 = 0) = P (X = 0) = 2
1
1 1 P (Y = 1) = P (X 2 = 1) = P (X = 1) = 2

Table for X, Y and P (Y = y)

X 1 1 1
E[Y ] = y · P (Y = y) = 0 · +1· =
y
2 2 2

and X
E[XY ] = X · Y · P (X = x, Y = y)
x,y

Since, Y = X 2 , the only possible pair (X, Y ) are (0, 0) and (1, 1)

X Y XY
0 0 0
1 1 1

Table for X,Y and XY

1
Thus, E[XY ] = 0 · P (X = 0) + 1 · P (X = 1) = 2
Now,

Cov(X, Y ) = E[XY ] − E[X]E[Y ]

  
1 1 1
= −
2 2 2
1 1
= −
2 4
= 0.25

12