UG18 - Nguyen Trinh Diem Quynh

Probability and Statistics

Instructor: Linh Tran Vinh

Probability and Statistics

Final synthesis

Spring 2021

I. Learning Questions
1. State the definitions to all concepts that are related to your topic
- My topic is the law of large number:
- Definition 1: A sample is a subset of the population.
- Definition 2: A random sample of size n from the distribution of X is a
collection
- of n independent random variables each with the same distribution as X
- Definition 3: An experiment is a set of positive outcomes that can be
repeated.
- Definition 4: A random variable, X, is a function that assigns to every
outcome of an experiment, a real numerical value. If X can assume at most
a finite or countably infinite number of possible values, X is said to be a
discrete random variable. If X can assume any value in some interval, or
intervals of real numbers and the probability that it assumes a specific given
value is 0, then X is said to be a continuous random variable
- Definition 5: The probability distribution of a discrete random variable X, is
a function f that assigns a probability to each potential value of X.
Additionally, f must satisfy the following necessary and sufficient conditions,
(1) f(x) = P(X = x) for x real, (2) f(x) ≥ 0, (3) P x f(x) = 1.
- Definition 6: The probability density function for a continuous random
variable X, denoted by f(x), is a function such that (1) f(x) ≥ 0, for all x in
+∞ 𝑏
R, (2) ∫ f(x)dx = 1, (3) P(a < X < b) = ∫f(x)dx for all a < b.
−∞ 𝑎
- Definition 7: Let X be a discrete random variable with density f(x), the

expected value or mean, denoted E(X) or µ, is µ = E(X) = ∑ 𝑥𝑓(𝑥)provided

𝑥

that ∑ 𝑥𝑓(𝑥) is finite

𝑥

1
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

- Definition 8: Let X be a random variable with mean µ, the variance of X

2 2 2 2 2 2
denoted Var(X) or σ is σ = Var(X) = E((X - µ) )= E(𝑋 ) −µ = E(𝑋 ) −
2
(E(X))
- Definition 9: The standard deviation of a random variable X, denoted σ is
the positive square root of the variance
- Definition 10: Let X1, X2, X3, ..., Xn be a random sample from the
𝑛 𝑥𝑖
distribution of X. The statistic ∑ 𝑛
is called the sample mean and is
𝑖=1
denoted X¯ n
- Definition 11: Random variables X1, X2, X3, ...Xn are said to be
independent and identically distributed or i.i.d. if each random variable Xi
has the same probability distribution as X1, and the occurrence of one does
not affect the probability of another.
- Definition 12: A continuous random variable X is said to follow a Poisson
−λ 𝑥
𝑒 λ
Distribution with mean and variance λ if its density f is given by f(x) = 𝑥!
for x = = 0, 1, 2, ... and λ > 0. We write X ∼ Pλ.
- Definition 13: A random variable Z is said to follow a standard normal
distribution if it has the following cumulative distribution function
2
𝑧 −𝑠
1 2
∫ 𝑒 𝑑𝑠
2π
−∞
- Definition 14: A Characteristic Function of a probability distribution is given
𝑖𝑡𝑋
by ϕX(t) = E(𝑒 ) where i = √ −1 and X is the variable with the given
distribution
- Definition 15: Chebyshev’s Inequality: with X be a random variable with
mean µ and standard deviation σ. Then for any positive number k P(| X −
2
σ
µ |≤ k) ≥ 1 − 2
𝑘
- Definition 16: The Central Limit Theorem: Let X1, X2, ..., Xn be a random
2
sample of size n from a distribution with mean µ and varianceσ . Then for
2
large n, X¯ is approximately normal with mean µ and variance σ /n.
Furthermore, for large n, the random variable (𝑋−µ)/( σ/√n) is
approximately standard normal.

2. State and prove (if the result was proved in the course) all related properties and
theorems
- State the Weak Law and prove it using Chebyshev’s Inequality
2
Assume that Var(𝑋𝑖) = σ for all i < ∞. Since the X1, X2, ...Xn are
independent, there is no correlation between them
𝑋1+𝑋2+𝑋3+...+𝑋𝑛
Therefore: Var(𝑋𝑛)= Var( 𝑛
)

2
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

2
= 1/𝑛 (Var(X1) + Var(X2) + ... + Var(Xn))
2 2
𝑛σ σ
= 2 = 𝑛
for n > 1
𝑛
Apply Chebyshev’s Inequality on X¯ n to get, for all ϵ > 0
2
σ
P(| 𝑋𝑛 − µ |> ) ≤ 2
𝑛ϵ
3. State and solve at least 2 applications / real-world problems related to the topic
In reality, the Law of Large Numbers has many useful applications to a variety of
fields. However, 2 applications that I choose to show are presented below:
a. How does the insurance company determine the effective prices for their
clients between the age of 23-25?
Answer: By analyzing and recording the number of accidents among people
aged 23-25, they can ascertain with a high degree of accuracy, the
probability of X amount of people aged 23 that will be the cause of an
accident in any given year. This allows them to set an effective price for
clients that fall into the range.

b. A survey of 1500 people is conducted to determine whether they prefer

ZaloPay or Momo. The results show that 27% of people prefer ZaloPay
while the remaining 73% favor Momo. Estimate the Margin of error in the
poll with a confidence of 90%.
Answer:
Let: 𝑋𝑛= 1 if n-th person is in favor of ZaloPay and equal to 0 otherwise
We have: P(X1 = 1) = 0.27, then:
2
µ = E(X1) = 0.27 and σ = 0.27 × 0.73 = 0.1971.
Therefore, with the law of large number, and n = 1500, we have:

(| 𝑋1+...𝑋𝑛
P |
| 1500
|
− 0. 27| ≥ ϵ ≤
| )
0.1971
1500ϵ
2 , ϵ> 0

1 0.1971
So if we set
10
= 2
1500ϵ

Then the ϵ = 0.036

II. Project/Essay Questions

1. Project’s name: Modeling Stock Price by Binomial and Gambler’s Ruin theory
a. Main goal:
- Our main goal is to build software by Python to help traders use it
as a reference to make better stock purchase decisions.

3
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

b. What were the concepts and techniques that you used in your
project/essay?
- The concept of Binomial Distribution and Gambler’s Ruin theory is
what we used in this project
- The Binomial Distribution is a probability distribution that
summarizes the likelihood that a value will take one of two
independent values under a given set of parameters. With n
is the number of trials, k is the number of successes desired,
p is the probability of getting success in one trial. And (1-p)
is the probability of getting a failure in one trial
- Gambler's ruin is a statistical concept, most commonly
expressed as the fact that a gambler playing a game with
negative expected value will eventually go broke, regardless
of their betting system. where player one wins each toss with
probability p, and player two wins with probability q = 1 − p

c. What did you learn from your group project/essay?

- My main takeaway from the group project was to improve my
cooperation abilities and learn how to use materials from the course
to real-life problems. In the beginning, the project has only two
members, which aids my coworker and me in efficiently
communicating and profoundly learning from each other's abilities
and expertise for developing the project. For example, I have a
background in finance, and my teammate is pursuing computer
science, so we can benefit from each other's experience while
determining the best strategy to complete the work. Besides, this
project helps me revise the class material and understand some
related theories while doing it.
d. What was the most interesting result? Why?
- The most intriguing part was seeing the binomial distribution graph
of our favorite stock, Facebook. We thought we could become
professional traders at the time. Because of that graph, we can see
the likelihood that the stock will rise in the future, which is crucial
information when investing. Even though we are aware that our
stock price prediction algorithm has many limitations, it is a step
closer for us to develop the stock price prediction model.

2. Assessment of group project

a. What worked well? What didn’t? Why?
- Our group project works pretty well most of the time. We meet once
or twice a week, and Messenger serves as our primary means of
contact. As a result, if a new problem arises, we fix it swiftly through

4
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

strange-forward communication. We distribute the responsibilities

evenly and always seek the assistance of our teammates whenever
we become overburdened. If we are unable to resolve an issue, we
call teachers for assistance.
b. What would you do differently going forward?
- Overcoming limitations and further developments are our next step
for moving forward,
- First, we must increase our understanding of the financial
industry, particularly the stock market, in order to further
enhance the model. Aside from daily stock price
measurements, we wish to improve our model to assess how
the market is influenced by events (such as a Covid or a
financial crisis), as they also have an impact on the stock
market.
- We must increase our understanding of user behavior. It took
some time for us to comprehend traders' behavior while
employing a price prediction tool. I believe our product only
partially meets their requirements. We wish to customize for
different types of traders because there are many different
types of merchants.
c. Are you proud of what your group produced? Why (not)?
- Yes, definitely, without a doubt. In the beginning, we all interested in
the topic that we choose, which is about stock market prediction.
Second, it takes us lots of time and effort to finish it, and we also
happy with the result turn out.
d. Was the group project experience useful? Why (not)?
- Yes, definitely. As previously said, it enhances the class information
that I have acquired, and this experience also helps me apply what I
have learned in a real-life setting. Furthermore, because this project
is being carried out in a small group, I am able to benefit from my
teammate's talents and additional information in order to contribute
to this project.
3. What is your assessment of your group member’s contribution to the project/essay?
How would you rate your group member’s contribution on a scale from 1-5 (lower
score means less great)
- I give him a 5 since he is a devoted and talented colleague. We
communicate really effectively, and as a result, we solve obstacles swiftly
when working. We are also compatible in terms of time management since
we are the first group to present the finished project with high-quality
material. Despite receiving comments to change some aspects of the project
after it was submitted, we still try our best in the end.

5
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

III. Course Innovation Questions

These two questions ask you to think about how best to improve our course.

4. What aspect(s) of the course worked well? Why?

- First and foremost, because the course is provided through the Minerva
platform, it is extremely user-friendly in terms of studying and engaging
with classmates, instructors, and class materials. Students have more
alternatives while interacting with others because of the multi-functionality
for expressing thoughts, and emotions. Furthermore, this platform features
a skilled technical support staff, as well as a well-designed, stylish, and
convenient platform. It would be fantastic if further Fulbright online
courses could be provided on this platform.
- Second, self-discovery is essential in learning, and it is also the most
effective approach for memorizing and comprehending the lesson. While
the course required a lot of cooperation and individual thinking, it
dramatically improved my learning experience. Despite the fact that it
caused greater discomfort for me to do, it was well worth the time and
effort at the end
- Third, because of the interesting youtube video, the break-time is
spectacular. This is the only class I've ever taken at Fulbright that had an
interesting break. Break-time is without a doubt my favorite portion of the
class. That is, I am learning not just in class, but also at breaks, through
funny and meaningful animated stories.
- Finally, one of the aspects of the course that I enjoy is the instructor's sense
of humor. This makes the course less difficult and motivates me to complete
the homework assignment.
5. What aspect(s) of the course is in most need of a change? Why?
I would consider this part as things that hinder my learning rather than what
should change.
- First, regarding the technological issue, the Minerva platform should have a
mechanism that notifies me of new assignments by Fulbright email, so that
I do not miss any assignments in this class. Furthermore, my classmates
and I were frequently having difficulty accessing material utilized in the
course. As a result, I propose that all of the class resources be neatly
organized in a file that is kept in the monitor bar on the Home Page. Those
resources should be carefully arranged so that students can easily find them
when they want to revise the class content.
- Second, consider the course's content. At first, I anticipate that the course
will demonstrate the use of probability and statistics in real-life situations.
However, it turned out to be a more calculation-driven course. As a result,
when we finish our final project, we are confident in our calculations but

6
UG18 - Nguyen Trinh Diem Quynh
Probability and Statistics
Instructor: Linh Tran Vinh

unsure whether anyone will utilize our product in their real-life experience
or not. Because I believe it is important to use correct concepts and theories
when building a product, but having actual customers experiencing and
showing the value of our product in their lives is a much more essential
factor when we evaluate the project.

IV. Self-evaluation Questions

6. What grade do you deserve? Why?

- I believe I am deserving of a B. Throughout the lesson, I actively engaged
and interacted. I am kind of a quiet person who seldom speaks in class, but
this class really motivates me to discuss and convey my opinions. Second, I
always turn in my homework and group projects on time (or even earlier
than the deadline). I prioritize this class over other classes this semester not
because I am worried about failing, but because I have fun and enjoy
studying and accomplishing assignments in this class. Despite the fact that I
seldom get the proper answer on the tests in this class. Furthermore, the
group presentation and final project are my favorite parts of the course, so
my teammate and I work extremely hard on them. So, though I may not
obtain a high mark in this class, I am really interested in and work hard on
it, which is why I believe a B is appropriate for me.