1. If he failed in chemistry, what is the probability that he failed in mathematics?

2. If he failed in mathematics, what is the probability that he failed in chemistry?

3. What is the probability that he failed mathematics or chemistry?

4. Is the event that the student has failed in chemistry is independent of the event that he has failed
in mathematics?
4

Solution:
Let M be the students who failed mathematics and C be the students who failed chemistry. Then

P (M ) = 0.25, P (C) = 0.15 and P (M ∩ C) = 0.10

1. The probability that a student failed mathematics, given that he has failed chemistry is
 
M P (M ∩ C) .10 2
P = = =
C P (C) .15 3

2. The probability that a student failed chemistry, given that he has failed mathematics is
 
C P (M ∩ C) .10 2
P = = =
M P (M ) .25 5

3. The probability that a student failed mathematics or chemistry is

P (M ∪ C) = P (M ) + P (C) − P (M ∩ C) = 0.25 + 0.15 − 0.10 = 0.30

Topic: Repeated Trials

Problem 3. Let X be the number of tosses for a fair coin up to and including the first time showing
heads. Compute the distribution function. 4

Solution:
1
The probability that heads will appear for the first time at the nth toss of a coin is 2n for n = 1, 2, . . .

These probabilities adds up to 1 i.e.

∞ ∞
X X 1
P ({X = n}) = =1
n=1 n=1
2n
The distribution function FX is given by

FX (x) = PX ((−∞, x]) = P ({X ∈ (−∞, x]}) (1)

X
= P ({X = n}) (2)
n<x
(
0 if x < 1
= 1 1 1
(3)
2 + 4 + ··· 2n if n ≤ x < n + 1, n = 1, 2, · · ·
(
0 if x < 1
= 1
(4)
1− 2n if n ≤ x < n + 1, n = 1, 2, · · ·

Topic: Random Variable

Problem 4. Consider rolling of a three faced dice with faces indicating H,T and S. Sample space is
given as Ω = {H,T,S}. The σ field is given as {{H,T}, {S}, {H,T,S}, {}}. Consider a function X : Ω → R
defined as

X(H) = 1, X(T) = 2, X(S) = 1.

2
Prove that X is not a random variable. Give another example when a function Y : Ω → R is not a
random variable. 4

Solution:
Given any set A in X, it must satisfy the probability law:

P (A) = P ({ω : X(ω) ∈ A})

This probability will be defined only if the set A belongs to σ field. Given the set A = 1, So the pre-image
of the set A is as follows

X −1 (A) = {ω : X(ω) = 1}
X −1 (A) = {H, S}
X −1 (A) ∈
/σ

So X is not a random variable.

Another example for function Y : Ω → R:

Y (H) = 1, Y (T) = 2, Y (S) = 2.

Using the argument similar to that for X, it can be proved that Y is not a random variable.

Topic: Discrete Random Variable

Problem 5. Let X be a discrete random variable with range RX = {1, 2, 3, ...}. Suppose the PMF
of X is given by
1
pX (k) =
2k
for k = 1, 2, 3, · · · .
Find and sketch the CDF of X, FX (x). Also, find P [2 < X ≤ 5] and P [X > 4]. 6

Solution:
bxc bxc
X X 1
FX (x) = P[X = i] = i
i=1 i=1
2
Hence,
P[2 < X ≤ 5] = P[X = 3] + P[X = 4] + P[X = 5] = 1/8 + 1/16 + 1/32 = 7/32
Alternative way
P[2 < X ≤ 5] = FX (5) − FX (2) = 1/8 + 1/16 + 1/32 = 7/32
P[X > 4] = 1 − FX (4) = 1 − (1/2 + 1/4 + 1/8 + 1/16) = 1/16

0.8

0.6
F X (x)

0.4

0.2

-1 0 1 2 3 4 5 6 7 8 9 10
x

Figure 1: CDF plot

3
 Topic: Continuous Random Variable

Problem 6. Let Ω = (0, 1) which the set of all real numbers between 0 and 1. Consider Borel sigma
algebra and uniform probability measure.
Let X(ω) be a random variable denoting the first digit of ω after decimal point. Compute the CDF of
X. Is it a discrete RV or continuous RV?
5

Solution:
x = −10, Bx = (−∞, −10], X −1 (Bx ) = φ PX [Bx ] = 0
x = 0, Bx = (−∞, 0], X −1 (B) = [0, .1), PX [Bx ] = 0.1
0 ≤ x < 1, Bx = (−∞, x], X −1 (B) = [0, .1), PX [Bx ] = 0.1
x = 1, Bx = (−∞, 1], X −1 (B) = [0, .2), PX [Bx ] = 0.2
1 ≤ x < 2, Bx = (−∞, x], X −1 (B) = [0, .2), PX [Bx ] = 0.2
...
x = 9, Bx = (−∞, 9], X −1 (B) = [0, 1), PX [Bx ] = 1
x > 9, Bx = (−∞, x], X −1 (B) = [0, 1), PX [Bx ] = 1
It takes only 10 values. It is a discrete RV.

Problem 7. Let us take a random experiment where a needle falls on the plane. The outcome is its
orientation (i.e. the angle with x axis). Ω = [0, π). Let us consider the standard sigma algebra and a
uniform measure.
Let X(Ω) be a random variable denoting the slop of the needle. Compute its CDF. What is the type of
RV? Compute its PDF or PMF (whichever is appropriate). 5

Solution:
Given that X : Ω → R is a random variable which represents the slope of the needle with respect to
x-axis. The uniform probability measure on Ω = [0, π) can be defined as:

x(ω) = tan ω for ω ∈ [0, π).

Note the ω ∈ [0, π) and tan−1 (a) ∈ (− π2 , π2 ).

20
X = tan ( )

15

10

5
)

0
X(

-5

-10

-15

-20
0 0.5 1 1.5 2 2.5 3

1. When x < 0.

FX (x) = P((−∞, x])

= P({ω : tan ω ∈ (−∞, x]})
π
= P({ω : < ω ≤ tan−1 (x) + π})
2
(tan−1 (x) + π) − ( π2 )
=
π−0
−1
tan (x) 1
= + .
π 2

4
 2. When x ≥ 0.

FX (x) = P((−∞, x])

= P({ω : tan ω ∈ (−∞, x]})
π
= P({ω : 0 < ω ≤ tan−1 (x) + })
2
(tan−1 (x) + π2 ) − (0)
=
π−0
tan−1 (x) 1
= + .
π 2

Combining above two cases, we get

tan−1 (x) 1
FX (x) = + for x ∈ (−∞, ∞).
π 2
Clearly, FX (x) is left continuous for all value of x ∈ (−∞, ∞), hence X is a continuous random variable.
Now, the PDF of X can be defined as:
d
fX (x) = FX (x)
dx 
d tan−1 (x) 1

= +
dx π 2
1
= .
π(1 + x2 )

Problem 8. Let Ω = [0, 1] with Borel sets and probability measure to be equal to Lebesgue measure.
Find P [X ∈ [0, 0.5)] for X(ω) = ω 2 . 5

Solution:
First we need to compute the inverse image
      
1 2 1 1
X ∈ 0, = ω ∈ [0, 1] : 0 ≤ ω ≤ = 0, √
2 2 2
h 
In the interval 0, √12 , under Lebesgue measure equal to its length:
    
1 1 1
P X ∈ 0, =P 0, =√
2 2 2
Topic: Practice Questions

Problem 9. Show that if X is constant function, then it is a random variable with respect to any
σ− field.

Solution:
Let B is the Borel set, c is the constant function. Consider two cases:
1. c ∈ B. Then {X ∈ B} = Ω.
2. c ∈
/ B. Then {X ∈ B} = φ.
But for every σ field F contains ω and φ. So X must be a random variable with respect to F.

Problem 10.
LetΩ = {1, 2, 3, 4} and the sigma-field F = {φ, ω, {1}, {2, 3, 4}}. Is X(ω) = 1 + ω is a random variable
with respect to F ? If not, give a counter example.

Solution:
For a function to be random variable, we need to find the inverse image of the function. If inverse
image present in the σ - field then we can conclude that the function is random variable. In this case
X(ω) = 1 + ω. Let us consider {X ∈ {3}} = {2} does not belong to F. X(ω) = 1 + ω is not a random

5
variable. For this case a function Y to be random variable, function must take a single value for {2, 3, 4}.
For example Y (2) = Y (3) = Y (4) = 1

Problem 11. For below functions, find the smallest sigma- field on Ω = {−2, −1, 0, 1, 2} with respect
to which the function is a random variable:
1. X(ω) = ω
2. X(ω) = 1 + ω

Solution:
For a function to be random variable, we need to ensure inverse image exists in the σ-field.
1. In this case ±a is mapped to a2 , so ±a should be made as single event. so the {0}, {−1, 1}, {−2, 2} are
mapped to 0, 1, 4 respectively.So the σ - field generated as F1 = {φ, Ω, {0}, {−1, 1}, {−2, 2}, {−2, −1, 1, 2},
{−2, 0, 2}, {−1, 0, 1}}
2. X(ω) = ω + 1, All the elemts of the Ω genrate
unique mapping in the real line, so the σ field will
consist of all subset of Ω i.e the Power set of Ω 2Ω .

Problem 12. From probability axioms, show that, If A1 , A2 , · · · An · · · ∈ A are disjoint subsets of X,
then !
n
[ n
X
P Ai = P (Ai )
i=1 i=1

Solution:
From third axiom of probability, we know that if Ai ’s are disjoint, then
∞ ∞
!
[ X
P An = P (An )
n=1 n=1

Now consider
S∞ Ai =
SNφ i > N . Note that Ai are still disjoint.
Hence i=1 Ai = i=1 Ai . Since P (Ai ) = 0 ∀i > N , so
∞
N
! !
[ [
P Ai =P Ai
i=1 i=1
∞
X
= P (Ai )
i=1
N
X ∞
X
= P (Ai ) + P (Ai ) (1)
i=1 i=N +1

Now Let us take Bi = Ai+N ∀i ≥ 1.

So
∞
X ∞
X
P (Al ) = P (Bl )
l=N +1 l=1
∞
!
[
=P Bl
l=1
= P(φ)
=0
Putting this back in eq (1) we get
∞
N
!
[ X
P Ai = P (Ai ) .
i=1 i=1