STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

Tutorial 1 Solutions
Exercise 1

a)

Firstly note that, 𝑝(1) = 𝑘, 𝑝(2) = 2𝑘, 𝑝(3) = 3𝑘, 𝑝(4) = 4𝑘, 𝑝(5) = 5𝑘 . Next, from Condition 2, it follows
that
1
𝑘 + 2𝑘 + 3𝑘 + 4𝑘 + 5𝑘 = 1. Hence, 15𝑘 = 1. Thus, 𝑘 = .
15

b)

Translating this statement into a probability expression, using the cdf notation, we arrive at the following:
1 6
𝑃(𝑋 ≤ 3) = 𝐹(3) = 𝑝(3) + 𝑝(2) + 𝑝(1), where 𝐹(3) = 6𝑘 = 6 (15) = 15.

c)
The keyword here is the phrase “inclusive”, taking care of that, we can re-express this word-statement and subsequently
arrive at the desired solution as follows:
1 9
𝑃(2 ≤ 𝑋 ≤ 4) = 𝑝(2) + 𝑝(3) + 𝑝(4) = 9𝑘 = 9 ( )= .
15 15
Cleverly, this can also be expressed in cdf notation using the proposition encountered in Lecture 1. This is achieved as
follows: 𝑃(2 ≤ 𝑋 ≤ 4) = 𝐹(4) − 𝐹(2− ) = 𝐹(4) − 𝐹(1). However, this constitutes no advantage since we have no
access to the full distribution function of X.

d)
1 4 9 16
Notice that, 𝑝(1) = , 𝑝(2) = , 𝑝(3) = , 𝑝(4) = , 𝑝(5) = 25. The first checkpoint is that each realization
50 50 50 50
assumes a positive probability; hence, Condition 1 is met. However, this alone does not conclude whether the distribution
is valid. Upon checking Condition 2, the subsequent conclusion can be made
1 55
(1 + 4 + 9 + 16 + 25) = = 1.1 ≠ 1
50 50
Thus, Condition 2 fails. Hence, this is not a valid probability mass function.

Exercise 2
a)
In this question, we mainly apply a counting technique (combinations) to answer the following questions. Finding the
domain space of X is effortless; it stems from the fact that we have two categories: defective (2) and non-defective (4). So,
when selecting three components, the possible values of X are 0 (when all three are non-defective), 1 (when only one is
defective and any 2 are non-defective), and 2 (when exactly 2 are defective and 1 is non-defective). Hence,

𝐷 = {0, 1, 2}.
STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

b)
Since the possible values that this random variable X assumes or takes on are countable (can be listed in a sequence), it
follows that X is a discrete random variable.
c)
To find the probability distribution of X, we use combinations, which follows the fact that listing all the outcomes in the
sample space is tedious; that is, we now use combinations as a shortcut. Hence,
(2𝐶0)(4𝐶3) 1 (2𝐶1)(4𝐶2) 3 (2𝐶2)(4𝐶1) 1
𝑝(0) = = , 𝑝(1) = = , 𝑝(2) = =
6𝐶3 5 6𝐶3 5 6𝐶3 5
d)
Expressing the pmf as a graph is very easy (left as a practical exercise). To express the pmf as a table, we proceed as
follows:
x 0 1 2
p(x) 0.2 0.6 0.2
e)
The least probable value is a realization of the random variable whose probability is the smallest when compared to the
probabilities associated with other possible values. Hence, in this case, the values 0 and 2 are the least probable realization
of X.

Exercise 3
a)
Suppose we have a set containing 5 unique elements, namely 1, 2, 3, 4, and 5, and we wish to select two elements in
which order does not matter. This is one way to visualize this problem. Before attempting to list all the simple events in
the sample space, we should worry about how many elements there are. To answer this, we use combinations, and we
obtain 𝑛(𝑆) = 5𝐶2 = 10.

When listing all the simple events in the sample space, we always respect the fact that it is a set, hence we should use the
proper notation: Hence, since this is a set, the order of elements does not matter (this not an order statistic ).
𝑆 = {(1,2), (1,3), (1,4), (1,5), (2,3), (2,4), (2,5), (3,4), (3,5), (4,5)}

b)
Organizing the information: 1 and 2 are defective. In this sample, we have 2 defective and 3 non-defective. This reminds
us of combinations. Firstly, our rule of association is defined as the number of defectives observed across the two selected
boards. We pause to apply counting techniques, and we simply map each outcome in S to a real number (deriving the
probability mass function from first principles, using the standard definition of a random variable).
Hence,
X ((1,2)) = 2 X ((2,3)) = 1 X ((3,5)) = 0
X ((1,3)) = 1 X ((2,4)) = 1 X ((4,5)) = 0
X ((1,4)) = 1 X ((2,5)) = 1
X ((1,5)) = 1 X ((3,4)) = 0
STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

Hence, the probability distribution of X follows immediately, and this is expressed as follows:
3 6 1
𝑝(0) = , 𝑝(1) = , 𝑝(2) =
10 10 10
The second approach uses combinations, and this is left as a practical exercise.

c)
The cumulative distribution function (cdf), which is sometimes called the distribution function is defined as follows:
𝐹(𝑥): ℝ → [0,1] for all 𝑥 ∈ ℝ
This is justified by the fact that 𝐹(𝑥) = 0 for all 𝑥 < 𝑥𝑚𝑖𝑛, and 𝐹(𝑥) = 1 for all 𝑥 ≥ 𝑥𝑚𝑎𝑥 .
d)
Note that, 𝑥𝑚𝑖𝑛 = 0 . Hence, 𝐹(0) = 𝑝(0) = 0.3. This is interpreted as follows:
The probability that there was no defective found among the two boards inspected is 30%.
Next, note that, 𝐹(1) ≠ 𝑝(1). But, 𝐹(1) = 𝑃(𝑋 ≤ 1) = 𝑝(1) + 𝑝(0) = 0.9. This means that the probability that at
most one defective will be observed in the two randomly selected boards is 90%.
Lastly, 𝑥𝑚𝑎𝑥 = 2, and thus, 𝐹(2) = 1. The interpretation is left as an exercise.

e)
Expressing 𝐹(𝑥) in graphical form as a step function can be done in Excel. Expressing this in functional form can also be
done in accordance with the prescribed book. Hence, this is an exercise left for students.

Exercise 4
a)
𝐷 = {0, 1, 2, 3}
b)
This is an easy exercise, a topic which was also covered in 1E. Hence, this is left as an exercise for students.
c)
To obtain this probability distribution, we need to use the Venn diagram. Firstly, N takes on 0 iff the selected individual
participates in none of the mentioned activities. The number of those individuals lies outside the three overlapping circles.
In the same way, we argue to understand other possible realizations of N. Thus, the probability distribution of N is given
as:
7 3 + 15 + 20 38 15 + 5 + 10 30 25
𝑝(0) = , 𝑝(1) = = , 𝑝(2) = = , 𝑝(3) =
100 100 100 100 100 100
d)
The probability mass function (pmf) is informally defined (proper scientific definition can be found in Lecture 1) as
follows:
STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

𝑝(𝑥): ℝ → [0,1] ∀ 𝑥 ∈ ℝ
The latter is true because 𝑝(𝑥) = 0 ∀ 𝑥 ∉ 𝐷 .

e)
The mode of a discrete random variable is its most likely realization, and in this case the possible value 2 has the highest
probability compared to other positive probabilities associated with other realizations of N.

Exercise 5
a)
This is trivial. Hence, this is left as an exercise for students.
b)

𝐷𝑌 = {8, 5, 4}
c)
To find the probability distribution of Y, we make use of Axiom 3 for probabilities. Firstly,
𝑃(𝑌 = 8) ⇔ 𝑃(𝑋 = −2) OR 𝑃(𝑋 = 2). The latter expressed formally, it reads as:
{𝑌 = 8} ⇔ {𝑋 = −2} ∪ {𝑋 = 2}. But the events {𝑋 = −2} AND {𝑋 = 2} are disjoint. So, from Axiom 3 in
probability theory, it follows that 𝑃({𝑋 = −2} ∪ {𝑋 = 2}) = 𝑃(𝑋 = −2) + 𝑃(𝑋 = 2) = 0.1 + 0.05 = 0.15.
Arguing in the same manner, we obtain the other probabilities:

𝑃(𝑌 = 5) = 0.6, 𝑃(4) = 0.25.

Hence, the probability mass function of Y is given as:
𝑝(8) = 0.15, 𝑝(5) = 0.6, 𝑝(4) = 0.25.
d)
𝑃(𝑌 ≠ 0) = 1 − 𝑃(𝑌 = 1) = 1 − 0.25 = 0.75 Complement rule
I made a mistake in this question (d & e), this should be calculated for the random variable X, as it carries little
significance for Y. Pretend that Y is X.

e)
Here, we apply the conditional probability formula, and that is achieved as follows:
𝑃({𝑌 = 2} ∩ {𝑌 = 0}) 0.05
𝑃(𝑌 = 2|𝑌 ≠ 0|) = = = 0.067
𝑃(𝑌 ≠ 0) 0.75

Exercise 6
a)
Applying Condition 2, we have:
STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

1
𝑟(|−3| + |−1| + |1| + |3|) = 8𝑟 = 1 ⟹ 𝑟 =
8

b)
One cool strategy to solve this problem would be to derive the distribution of 𝑋 2 , however that might be tedious
(for additional exercise please try it). As a shortcut, we will solve this inequality, keeping in mind the properties
of a cdf. Upon doing the latter, we obtain the following probability expression:
3 4 7
𝑃({𝑋 ≤ −0.6633} ∪ 𝑋{𝑋 ≥ 0.6633}) = 𝐹(−3) + [1 − 𝐹(−1)] = + [1 − ( )] = .
8 8 8
c)
Firstly, we need to find the domain of Y, which is given as: 𝐷𝑌 = {−20, −2, −8,2}. Hence, the pmf of Y is,
3 1 1 3
𝑝(−20) = , 𝑝(−2) = , 𝑝(−8) = , 𝑝(2) =
8 8 8 8
Exercise 7
a)
𝑑 = 1, since F(x_max)=1.
b)
Since the possible values that X assumes or takes are strictly positive real numbers, it follows that X is a non-
negative random variable.
c)
Value of 𝑎: Note that, 𝑎 = 𝑝(2) = 𝐹(2) − 𝐹(1) = 0.26 − 0.10 = 0.16.

Value of 𝑏 : In an almost similar way, but still using the cdf notation we can find the value of b as follows:
𝑏 = 𝐹(3) = 𝑝(3) + 𝑝(2) + 𝑝(1) = 0.10 + 0.16 + 0.28 = 0.54.
Lastly, the value of 𝑐, can be found as follows:
𝑐 = 𝑝(4) = 𝐹(4) − 𝐹(3) = 0.76 − 0.54 = 0.22
d)
The solution provided below assumes that this selection is done without replacement (for additional exercise,
the students should assume that it is done with replacement):
To visualize this, suppose that you have a collection of distinct elements, namely 1, 2, 3, 4, and 5. To obtain odd
numbers, our favorable outcomes are the following: The order is important here, because the selections are
independent.
(1,3) OR (3,1) OR (3,5) OR (5,3) OR (5,1) OR (1,5)
Hence, the desire probability is (this a cool application of Axiom 3 and Independence):
STA2EA1 (University of Johannesburg) Distribution Theory Part 2 2026

𝑃(both odd) = 2[0.1 × 0.28 + 0.28 × 0.24 +× 0.24 × 0.1] = 0.2384

e)
Conditional Probability, Independence, and Axiom 3 allow the following sequence of arguments:
𝑃({(1,5)} ∪ {5,1}) 2 × 0.24 × 0.10
𝑃( {sum is 6} ∣ {both odd } ) = = = 0.2011
𝑃(both odd) 0.2384
Exercise 8
a)
This is a nice application of the cdf proposition encountered in Lecture 1. Since, 𝑃(𝑋 > 𝑛), and 𝑋 is discrete, it follows
that the next possible value in the sequence of realizations of X is included. Hence, 𝑃(𝑋 ≥ 𝑛 + 1). From this, we can
think of a complement rule as the preceding probability is not in the cdf form, to that end, we obtain the following re-
expression: 𝑃(𝑋 ≥ 𝑛 + 1) = 1 − 𝑃(𝑋 ≤ (𝑛 + 1)− ) = 1 − 𝑃(𝑋 ≤ 𝑛) = 1 − 𝐹(𝑛) = 𝑎 𝑛 ⇒ 𝐹(𝑛) = 1 − 𝑎𝑛 .

Thus, 𝐹(𝑛) = 1 − 𝑎𝑛 ∀ 𝑛 = 0, 1,2, 3, ….

b)

𝐹(0) = 1 − 𝑎0 = 1.
c & e) Left as an exercise.
d)
Recall, that 𝑃(𝑋 = 𝑥) = 𝐹(𝑥) − 𝐹(𝑥 − 1). Hence,
𝑝(𝑛) = 𝑎𝑛−1 (𝑎 − 1) for 𝑛 = 1, 2, 3, …. Simplifying this, we obtain the following:

Exercise 9
a)
Since, 0.5 ∉ 𝐷 it follows that 𝑝(0.5) = 0.
b)
1
From Condition2, it follows that, 3𝑟 + 3𝑠 = 1 ⇔ 𝑟 + 𝑠 = 6. To solve for these two unknowns, we utilize the
extra information given: 3𝑠 = 3(3𝑟) ⇔ 𝑠 = 3𝑟.
I made a mistake that cdf relation should be, 𝑃(𝑋 > 2) = 3𝑃(𝑋 ≤ 2). Solving simultaneously, we get:
1 1
𝑟= , 𝑠= .
24 8
c, d, & e) Left as an exercise.

Exercise 10-12 (Left as an exercise: solutions can be found in class, in the tutor venue or from the Lecturer)

******************************** END-OF-TUTORIAL 1 SOLUTIONS ************************