Faculty of Engineering and Natural Sciences

Electrical Engineering and Electronics

Introduction to Probability and Statistics

HOMEWORK 1#

Student: Muhamed Selimović

Professor: PhD. Lejla Miller

Student ID: 220302103

Date: 10th January, 2025

 1. State and prove Chebyshev’s inequality for continuous distributions. Give an example to
illustrate it, using some continuous distribution.

Chebyshev’s inequality tells us that, no matter the shape of the distribution (it could be normal,
uniform, skewed, or anything else), a certain percentage of values will always lie within a
specific range around the mean. Let X be a random variable with a finite mean denoted as µ and
a finite non-zero variance, which is denoted as 𝜎 2 , for any real number t > 0, here states:
𝜎2
P (∣X−μ∣ ≥ t) ≤
𝑡2

If we set t = 10, Chebyshev's inequality tells us that at least:

1 1
P (∣X−μ∣ < 10σ) ≥ 1 − =1− = 0.99
102 100
This means that at least 99% of the data /cases will lie within 10 standard deviations
from the mean.

To prove this, we can go like this:

P (∣X−μ∣ ≥ t) = P (∣ 𝑋 − 𝜇 ∣2 ) ≥ 𝑡 2 Since the square gives non-negative value, we can apply
Markov`s inequality:
𝐸 (∣𝑋−𝜇∣2 )
P (∣ 𝑋 − 𝜇 ∣2 ) ≥ 𝑡 2 ≤ and therefore 𝐸 (∣ 𝑋 − 𝜇 ∣2 ) is the variance of x (standard
𝑡2
deviation to the power of 2).The difference between X and mean is somehow limited
by variance of x, Var(X). This is intuitively expected as variance shows on average how far we
are from the mean.

2. Define multinomial discrete distribution, give some basic facts and give 2 examples.

Multinomial discrete distribution is a generalization of standard binomial distribution. In this

kind of distribution, the number of possible outcomes on any given trial is allowed to be
more than 2.

Example 1:
Imagine an election where 100 people are voting for one of three political candidates:
Candidate A, Candidate B, and Candidate C. The probability of a voter choosing each
candidate is prescribed in this way:
 • Candidate A: (0.4)
• Candidate B: (0.35)
• Candidate C: (0.25)
The number of votes each candidate receives can be modeled as a multinomial distribution,
where you want to find the probability of a specific outcome, such as Candidate A getting 45
votes, Candidate B getting 35 votes, and Candidate C getting 20 votes, out of the total 100
votes cast. We`ll use formula for this example:
100!
P(𝑋𝐴 = 45, 𝑋𝐵 = 35, 𝑋𝐶 = 20) = 0.445 0.3535 0.2525
45!35!20!
Example 2:

Let's calculate the probability of selling 50 units of Product X, 30 units of Product Y, and 20
units of Product Z in the North region. The probabilities for the products are:

• Px = 0.5
• Py = 0.3
• Pz = 0.2

n = 100

The number of units sold for each product: Xx = 50, Xy = 30, Xz = 20

100!
P = 50!30!20! 0.550 0.330 0.220

3. Derive the mean and variance for:

a) Exponential distribution
b) Poisson distribution

In Poisson distribution, the mean of the distribution is represented by λ and e is

constant, which is approximately equal to 2.71828. Then, the Poisson probability is:

P(x, λ ) =(e– λ λx)/x!

In Poisson distribution, the mean is represented as E(X) = λ.

For a Poisson Distribution, the mean and the variance are equal. It means that E(X) =
Var(X)

Where,
 Var(X) is the variance as it is explained earlier.

c) Uniform distribution

Mean of the Uniform Distribution = (a+b) / 2

The variance of the uniform distribution is:
σ2 = b-a2 / 12
The density function, here, is:
F(x) = 1 / (b-a)

4. Suppose X and Y are discrete finite random variables. Show that E(X+Y) = E(X) + E(Y):

5. Suppose X and Y and discrete finite random variables

a) Show if X,Y are independent, then E(XY) = E(X)E(Y)
 b) When is E(X2) = (E(X))2? Explain.
It is not the case generally because square of the X cases gives only positive values
therefore the expected value of squared X is higher than the square of mean value for
that same X. However, E(𝑋 2 ) = (E(X))2 is correct when the random variable X takes a
constant value. In other words, if X is a deterministic random variable ( always takes the same
value with probability 1, then:
E(X)=c for some constant c,
E(𝑋 2 ) = 𝑐 2
Therefore, E(𝑋 2 ) = (E(X))2 = 𝑐 2
In this case the variance is zero because there is no spread in the values that X takes.

6. Give an example of a continuous random variable X (by giving its PDF and proving) so
that:
a) E(X) doesn’t exist i.e. diverges
 For a continuous random variable X, the expected value E(X)) is given by:
∞
𝐸(𝑋) = ∫ 𝑥𝑓(𝑥)𝑑𝑥
−∞
This integral must converge for E(X) to exist. However, if the integral diverges, then E(X)
does not exist.
The Cauchy distribution is a continuous random variable with the probability density
function (PDF):

Thus, the integral does not converge, and the expected value E(X) does not exist for the
Cauchy distribution.

b) E(X) is finite but Var(X) doesn’t exist i.e. diverges:

 7.
8.
9.
10.
11.
12.

7. In class we learned how to compute confidence intervals for population means in the
case that the standard deviation is known. Often, standard deviation is unknown and
confidence intervals are based on sample mean and sample standard deviation.
Describe the method/formula used, and give two examples.:

If the standard deviation is unknown we use the t-distribution, also known as Student’s t-
distribution, is a way of describing data that follow a bell curve when plotted on a graph, with
the greatest number of observations close to the mean and fewer observations in the tails.It is
a type of normal distribution used for smaller sample sizes, where the variance in the data is
unknown.The T-Distribution is a measure of probability (p-value). It is used to find the statistical
significance when the sample size is small, i.e., less than 30, with an obscure standard
deviation. The mean of a
T-Distribution is evaluated as zero, and the variance is derived as v/(v-2), where v is the degree
of freedom.This method finds extreme values—the confidence interval's lower and upper
limits.
Example 1: ABC Poultry Farms supplies eggs. The company claims its eggs remain fresh for five
days if refrigerated. An analyst samples 25 eggs to test this claim. The average freshness of eggs
was 4.5 days, with a standard deviation of a day. If the company's claim is true, find the
probability of all selected eggs lasting about 4.5 days.

• x̄ = 4.5 days
• μ = 5 days
• s = 1 day
• n = 25
• Therefore,
t = (x̄-µ)/(s/√n)
• t = (4.5 – 5)/(1/√25)
• t = -0.5/0.2 = -2.5
• Since the minus sign is irrelevant here, we get t = 2.5.
• Degree of Freedom (df) = n – 1
• df = 25 – 1 = 24
• Thus, according to the t-test, the probability (p-value) of eggs not lasting for more
than 4.5 days is 0.01965418.

Example 2: A random sample of statistics students were asked to estimate the total number of
hours they spend watching television in an average week. The responses are recorded in this
table. Use this sample data to construct a 98% confidence interval for the mean number of
hours statistics students will spend watching television in one week.

0 3 1 20

5 10 1 10

14 2 4 4
x = 6.133, s = 5.514, n = 15, and df = 15-1=14

CL = 0.98, so α = 1 – CL = 1 - 0.98 = 0.02

𝑡𝛼 = 𝑡0.01 = 2.624
2

𝑠 5.514
EBM = 𝑡𝛼 ( 𝑛) = 2.624 ( ) = 3.736
2 √ √15

X – EBM = 2.397 X + EBM = 9.869 It`s 98% shure that mean of students spending time on
TV is between two of these numbers.