Introduction

Tree Diagrams are one of the ways that are used to solve Bayes' rule problems. Bayes rule
is an equation that helps in extending the use of the law of conditional probabilities. It helps
in relating or revising original probabilities with the new information. Probabilities are the
chances of a thing happening. It is a numerical representation of chances. Its value is always
between 1 and 0 where "1" denotes 100 percent chances and "0" denotes 0 percent
chances. A tree diagram of probabilities shows a drawing that represents prior probabilities
and their relation to conditional probabilities. Tree diagrams help in understanding the
relationship between the prior probability and the new information that is provided in an easy
way. It simplifies the complex equation with the help of a diagram. The tree diagrams help in
the decision-making process over a complex event that can have several outcomes.

Conditional Probability
Conditional probability is a dependent probability that is depending upon some other
probability. In simple words, the probability that has dependent on other factors and can only
be seen when some conditional factors take place is called conditional probability.
Conditional probability is calculated if prior information on the relevant probability is pre-
studied. If A and B are two events, where A's conditional probability is to be found out and it
is known that B is known or has occurred. It can be done by a simple formula,

𝑃 (𝐴 ∩ 𝐵) 𝑃(𝐴). 𝑃(𝐵|𝐴)
𝑃(𝐴|𝐵) = =
𝑃 (𝐵) 𝑃(𝐵)

Where
P (A∩B) is the probability of both A and B events having together,
P (A|B) is the Conditional Probability of event A,
P (B) is for the Probability of event B,
P (A) is the probability of event A,
P (B|A) is the Conditional Probability of event B (must be given).
In order to simplify the calculation and easy for understanding there is one more method that
can help in understanding and representing conditional probability and that method is called
Tree Diagram. This method is an extension of Bayes' rule for solving conditional
probabilities. Bayes' rule was first introduced from 1702 to 1761 by Thomas Bayes. In his
rules, he forms an equation to solve Conditional Probability. This equation can be solved in
a form of a diagram that is called a Tree Diagram as the name suggested this diagram is in a
form of a tree, as a tree has different branches this diagram also has, the branches grow
from left to right and each branch represents either a probability or a piece of new
information. The further left side shows the original probability and when we move towards
the right expanding branches it shows new information. In order to show the relationship
between branches or we can say between the events digits are multiplied to show their
future probabilities.

Probability Tree Diagram (PART 1)

In our case study, the following information is provided and was told to make a Probability
Tree Diagram. We were told there are three flights namely "Amira", "Biyas" and "Chinar"
operating in Srinagar with different serving capacities. They each have different on-time
ratings also. Serving capacity and on-time rating can be found in table below,

Flight Name Serving Capacity On-Time Rating

Amira 50% 80%
Biyas 30% 65%
Chinar 20% 40%

Solution:-
Conditional Probability for Flight “Amira” (PART 2)

Conditional probability is an interdependent set of events whose chances of happening

depend on others even what is prior known. These probabilities are depending on other
events and another event also depends on it. In our case study three flights "Amira", "Biyas"
and "Chinar" have serving capacities of 50%, 30%, and 20% respectively. Their on-time
rating is 80%, 65%, and 40% respectively. Now the Conditional Probability of a flight that left
and is "Amira" can be calculated as follow,

Let as assume that,

A, can help in represent the scheduled flights for airline “Amira”.

B, can help in represent the scheduled flights for airline “Biyas”.
C, can help in represent the scheduled flights for airline “Chinar”.

The probability of the flights is:

P (Amira) = P(A) = 0.50

P (Biyas) = P(B) = 0.30
P (Chinar) = P(C) = 0.20

Let us consider the event of “flight on-time” be E.

Conditional Provability for "Amira", "Biyas" and "Chinar" for the event E (flight on-time).

P (Event flight on-time |Amira) = P(E|A) = 0.80

P (Event flight on-time |Biyas) = P(E|B) = 0.65
P (Event flight on-time |Chinar) = P(E|C) = 0.40

We have given data and now we have to figure out that the plane just left was of “Amira”.
That means known events are the flight must be of “Amira” and it should be on time out of all
the flights that are on time. Let this be P(A|E)
According to the equation of conditional Probability:-

𝑃 (𝐴). 𝑃(𝐸|𝐴)
𝑃(𝐴|𝐸 ) =
𝑃(𝐴). 𝑃(𝐸 |𝐴) + 𝑃 (𝐵). 𝑃 (𝐸|𝐵) + 𝑃(𝐶 ). 𝑃(𝐸|𝐶)

(. 50) × (.80)
𝑃 (𝐴|𝐸) =
(. 50) × (. 80) + (. 30) × (.65) + (. 20) × (.40)

(.40)
𝑃 (𝐴|𝐸) =
(. 40) + (. 195) + (. 08)

(. 40)
𝑃(𝐴|𝐸 =
(.675)

𝑷(𝑨|𝑬) = . 𝟓𝟗𝟐𝟓𝟗

Thus as per our calculation probability of “Amira” that just left from Srinagar on-time is .5926
or in term of percentage its 59.26%

Conclusion

A probability tree diagram is very useful for solving complex problems and it's very easy to
understand. When there is a requirement for solving a complex and mixed probability for
events that are interlinked and interdependent the probability tree diagram is used. It helps in
making a graphical representation of data in a form of a diagram that help is mapping out
opportunity and information provided cases. In some cases, Conditional Probability is hard to
understand and hard to predict the outcome as it requires a complex formula that keeps on
growing if events are added. Once the equation is solved it's hard to make changes but in a
tree diagram, it is easy to understand and work with. In some cases additional information or
events are added it can be solved but just expanding its branches.
Introduction
Variables are the factors whose values keep on changing with the required information.
Variables are the very crucial factors in statistics as they help with the derivation of
information and having a quick change whenever new information is required. They are
helpful in determining the features of various values and can help in solving research and
finding definitive values. Variables can be placed in a calculation for the values that can be
changed in the future such as salary, temperature, sales, etc. In short, we can say that
variables help to establish a correlation between different components of an observation. An
equation or Observation can have one or more variables in it. The variables in regression
analysis are Dependent variables and Independent variables and help in the observation of
how one variable is affected by the desired change in other. Regression Analysis is a
modern-day tool that helps in forecasting the future even when there is a change in any one
or more variables.

Variables and Its types (PART 1)

Variables are the key factors of observation that help in presenting the relationship in
different elements of data. All the elements of a data sheet that affect the result can be
called variables. Variables are the data representation and they help information of
equations that can help in the calculation of the relationship between one value to another. A
variable can be in a form of numeric or alphabetic but it should be relevant to the equation or
observation. There are two variables in common, Independent and Dependent variables.

Dependent variables:- When we observe data there can be a relationship between

different variables. When a variable is dependent on another variable it is called the
dependent variable. In simple words, a dependent variable is a variable whose value can be
calculated with the help of other variables. In Regression Analysis the variable to be
predicted is called the dependent variable and it is denoted by “y”.

Independent variables:- When a variable is independent of another variable it is called

the independent variable. An independent variable also called explanatory variable is a
variable whose value is used to calculate the value of other variables. It helps in calculating
the dependency of dependent variables. In Regression Analysis the variable that acts as a
predictor is called the independent variable and it is denoted by x.
Regression Model and its Equation (PART 2)

Regression Analysis is a process of constructing a mathematical model or function that is

used to predict or determine the changes in one variable by changes in another variable.
There can use to predict one variable with the help of other variables. It is used for two main
reasons to measure the influence on one variable by one or more other variables and
prediction of one variable by one or more variables. There can be Simple Linear Regression
or Multiple Linear Regression. In Simple Linear Regression, there are only two variables one
that is Dependent and one Independent, in short, it has only one independent variable to
affect the Dependent variable. In Multiple Linear Regression, there can be more than one
independent variable that affects the value or dependent variable.

In order to predict the accuracy and observe the dependency of variables, it is important to
back up regression analysis with an appropriate mathematical equation. The following
mathematically equation can help in predicting the future or measuring the current impact on
the dependent variable by independent variables:-

𝒚 = 𝒙𝜷𝟎 + 𝜷𝟏 + 𝝐𝒊

Where,

y = Dependent variable,
x = Independent variable,
𝝐𝒊 = error of prediction,
𝜷𝟎 = slope of the line,
𝜷𝟏 = intercept of the line.

Note

“𝛽1 ” can be calculated by

𝑵∑𝒙𝒚 − (∑𝒙)(∑𝒚)
𝜷𝟏 =
𝑵∑𝒙𝟐 − (∑𝒙)𝟐

“𝛽0 ” can be calculated by

∑𝒚 − 𝜷𝟏 ∑𝒙
𝜷𝟎 =
𝑵
Regression Analysis In EXCEL for given data (PART 3)
In our case scenario, Rashmi Dhar wants to know the effect of her spending on the
Advertisement, the Number of Salespeople, and the Customer rating of her "Kahwa" on the
sale. The DATA provided is as bellow:-

Output from the data regression analysis in Excel

Interpretation of Regression Statistics-Table, ANOVA-Table
(PART 4)
Multiple R :- Multiple R measures the relationship between the independent variable and
dependent variable. In our analysis, it is 0.98 or 98% which is quite high.

R Square :- R Square helps us to present the accuracy of our calculation in our analysis it is
0.96 or 96% which is also quite high.

Signification f :- Signification f shows the importance of the independent variables to the

dependent variable. This analysis shows how the values of advertisement, salespeople, and
ratings affect sales. The numeric value is 0.000113414 and that is less than 0.05 which
shows they are significant to sales.

p-values:- It is the same as signification f but at an individual level. Same like signification fit
follows the same rule that if it is less than 0.05 then it is significant. In our case, the rating
that is 0.004730 is less than 0.05 which is significant but Advertisement i.e. 0.303811 and
Salesperson i.e. 0.888322 are not significant.

Coefficients:- Coefficients helps in deriving the formula that helps in predicting the sales. By
using the formula y= mx + b. In our analysis, we find out three independent variables are
used do generate the formula.

t-stat:- It is used for finding t test in analysis.

ANOVA Table:- It shows significance f and also k values together with total that is n-1 and
residual that is n-k-1

Conclusion
The Regression analysis is a systematic analysis of data that can help in forecasting the
future prediction and measuring the effects of current values of independent variables on
dependent variables. In the case study, we can see that the value of dependent variables is
influenced by the independent variables and can help in forecasting the future sales if there
are any changes made in independent variables. Future sales expansion can be done by an
optimal change in the independent factors that are advertisement, rating, and salesperson.
Introduction
Probability is a system of calculating the chances of an outcome that can happen for an
event. It is a very important part of statistic management and helps in forming strategies for
the future. It helps in predicting the future result with the help of given data. All Business
performs surveys and then analyzes data to run a probability test to check the chances of
events that can take place. Instagram introduces REELS as their new feature. Some
probability in which we have to obtain 'x' outcome in 'n' trail can be solved by Binomial
Formula. A binomial problem can be solved by an equation or can be simplified with the help
of a Binomial table.

Binomial Distribution
Before knowing a Binomial distribution we have to understand what is Discrete Distribution.
A Discrete Distribution is a random variable with a set of all possible values that must be
finite or a countable infinite number of possible values. For all the discrete distributions
Binomial Distribution is widely known. Binomial distributions are solved with the help of an
equation and that is called a binomial formula.

Binomial Formula

𝒏!
𝑷(𝒙) = 𝒏𝑪𝒙 × 𝒑𝒙 × 𝒒𝒏−𝒙 = × 𝒑𝒙 × 𝒒𝒏−𝒙
𝒙! (𝒏 − 𝒙)!

Where,

n = Number of Trials or Number of Sample,

x = Number of Successes Desired,
p = Probability of getting Success in One Trials,
q = (1 – p) Probability of getting a Failure in One Trials.
Case Study on Instagram REELS
The survey that was conducted in India resulted that there are 75% of Instagram users who
loves REELS. A random approach is done in the University in Vile Parle with 25 Instagram
Users.

Data provided in the Statements is,

n = Number of Sample. = 25
p = Probability of getting Success in One Trials. = 75% or 0.75
q = Probability of getting a Failure in One Trials. (1 – p) = (1-0.75) = 0.25
x = Number of Successes Desired = ‘a’ 15 and ‘b’ 20

As per the Binomial formula which will be used in solving following problems.

𝑷(𝒙) = 𝒏𝑪𝒙 × 𝒑𝒙 × 𝒒𝒏−𝒙

Solution A

Probability to show 15 of them like REELS so we can say x = 15

𝑷(𝟏𝟓) = 𝟐𝟓𝑪𝟏𝟓 × 𝟎. 𝟕𝟓𝟏𝟓 × 𝟎. 𝟐𝟓𝟐𝟓−𝟏𝟓

𝑷(𝟏𝟓) = 𝟑𝟐𝟔𝟖𝟕𝟔𝟎 × 𝟎. 𝟎𝟏𝟑𝟑𝟔𝟑𝟒𝟔 × 𝟎. 𝟎𝟎𝟎𝟎𝟎𝟎𝟗𝟓𝟑𝟔𝟕𝟒

𝑷(𝟏𝟓) = 0.04165

Solution B

Probability to shows 20 of them like REELS so we can say x = 20

𝑷(𝟐𝟎) = 𝟐𝟓𝑪𝟐𝟎 × 𝟎. 𝟕𝟓𝟐𝟎 × 𝟎. 𝟐𝟓𝟐𝟓−𝟐𝟎

𝑷(𝟐𝟎) = 𝟓𝟑𝟏𝟑𝟎 × 𝟎. 𝟎𝟎𝟑𝟏𝟕𝟏𝟐𝟏 × 𝟎. 𝟎𝟎𝟎𝟗𝟕𝟔𝟓𝟔𝟑

𝑷(𝟐𝟎) = 0.16453
Conclusion
The binomial formula helps in solving the prediction of situations given for specific data. In
our case study, Instagram REELS is a new feature and 75% of Indians liked it. There was a
random survey done on 25 people and there were two cases in the first case was what
chances are 15 out of 25 likes REELS and the second was what chances are 20 out of 25
likes REELS. This can be solved by Binomial Formula and probability can find out in both the
cases.

Introduction
Statistics is a branch of management that analysis the current output and helps in predicting
future results by performing experiments. It statistics experiment involves can have random
outcomes. Results can be called distribution and they are of two types Discrete and
Continuous distribution. In our Case study about "Bharatdashan", the mean and standard
deviation is given and we have to find the probability of less than or greater than of mean
value. This comes under Normal Distribution and can be solved by an equation or by a
graph.

Normal Distribution
Before understanding Normal Distribution we have to understand Continuous Distribution. A
continuous distribution is based on continuous variables which means, they are based on
variables taking on a value on every point over a given interval. Normal Distribution is a part
of Continuous Distribution. These problems can be solved by an equation, and it is also
called “z” formula

Normal Distribution Formula (z)

 𝒙−𝝁
𝒛= , 𝝈 ≠𝟎
𝝈
Where,

µ = Mean of ‘x’
𝜎 = Standard Deviation of ‘x’
z = Probability of getting ‘x’
x = Value of Given Normal Distribution.

Case Study on “Bharatdarshan”

In our case study, the number of daily visits hit is a normal distribution of random variables
with a mean of 10,000 and a Standard Deviation of 2,400. We have to find out two
probability, a) if there is more than 12,000 hits and b) if there is less than 9,000 hits.

Data provided in the Statements is,

µ = Mean of ‘x’ = 10,000

𝜎 = Standard Deviation of ‘x’ = 2,400
x = Value of Given Normal Distribution. = X1 (12,000), X2 (9,000)

Before solving for answers we have to first understand that a Normal Distribution is in a form
of a bell curved graph that is highest in the middle and slopes downward from both sides till
it touches the base. The highest point at the centre is called Mean and its right ride is X
where the left side is -X. The area under the curve is denoted by 1. The outcome result is
run through the Z table to calculate the correct answer. If the X is greater than, then it must
be subtracted from 1.

Solution A
Probability of getting more than 12,000 hits
𝒙−𝝁
𝒛= , 𝝈 ≠𝟎
𝝈
𝟏𝟐, 𝟎𝟎𝟎 − 𝟏𝟎, 𝟎𝟎𝟎
𝒛=
𝟐, 𝟒𝟎𝟎

𝟐, 𝟎𝟎𝟎
𝒛=
𝟐, 𝟒𝟎𝟎

𝒛 = 𝟎. 𝟖𝟑𝟑𝟑𝟑 "𝒐𝒓" 𝟎. 𝟖𝟑

Using Z table

When z = 0.83 then value of p will be 0.7967 but since X > mean so it must be
subtracted from 1

𝑷 = 𝟏 − 𝟎. 𝟕𝟗𝟔𝟕

𝑷 = 𝟎. 𝟐𝟎𝟑𝟑 ′𝒐𝒓′ 𝟐𝟎. 𝟑𝟑%

Solution B
Probability of getting less than 9,000 hits
𝒙−𝝁
𝒛= , 𝝈 ≠𝟎
𝝈
𝟗, 𝟎𝟎𝟎 − 𝟏𝟎, 𝟎𝟎𝟎
𝒛=
𝟐, 𝟒𝟎𝟎

𝟏, 𝟎𝟎𝟎
𝒛=
𝟐, 𝟒𝟎𝟎

𝒛 = −𝟎. 𝟒𝟏𝟔𝟔 "or" − 𝟎. 𝟒𝟐

Using Z table

𝑷 = 𝟎. 𝟑𝟑𝟕𝟐 ′𝒐𝒓′ 𝟑𝟑. 𝟕𝟐%

Conclusion
When there are problems that are in a continuous distribution solutions to the problems can
find out by the Normal Distribution method. Since it is a continuous variable it is easier to find
out results in graphical format. In our case scenarios, there is a chance of 20.33% in getting
hits more than 12,000, and 33.72% in getting hits less than 9,000.