DECISION SCIENCE

ANS-1
Introduction:
Business decision-making involves risks because it has certain uncertainties
attached along. In business, the importance of probability in business decision-
making is all about looking into events. Probability simply means “possibility”. This
can be also seen as the events that are likely to occur.
Concept & application:
As already we know that probability is the measure of the likelihood for the events to
take place for example tossing up a coin can have two possible outcomes
Either a Head or a Tail
In this tossing up of the coin is the Event taking place and dropping down as a Head
or a Tail is the possible outcome of the event. When the coin is flipped we do not
know what shall be the outcome but we know that it will be a head or a tail i.e there
can be 2 possible outcomes
Hence the probability of a head or a tail will be ½
Note:
However, it is a point to be noted that the total of all the probabilities for all the
events should be equal to “zero”.
Mathematically, probability can be calculated in the following way
Probability of an event = no. of ways an event can occur/ total no. of events
There can be probability based on the nature of the outcomes or the approach
followed while locating the probability of an event occurring.
There can be four types of probabilities namely as below
1. Marginal probability:
This is found by diving the subtotal by the whole total.

2. Union probability:
Under this probability for example there are two events E1 & E2, either E1 will
occur or E2 will occur, or both E1&E2 will occur.

3. Joint probability;
Under this probability both the events, E1&E2 must occur.

4. Conditional probability:
Under this type of probability, the probability of E1 occurring will depend on
the fact that event E2 has already occurred.
The mentioned problem is based on the concept of conditional probability concept.
The conditional probability is based on the knowledge of one of the two events that
are being studied. Conditional probability can be understood as the possibility of an
event or outcome happening based on the existence of previous events or
outcomes.
Definition and formulae:
By defining, the probability of the occurrence of an event A when another event B is
about event A has already occurred is known as the conditional probability

If X and Y are the two events, the conditional probability of X occurring is given that
Y is known or has already occurred.
The concept is one of the quintessential concepts in probability theory. however
conditional probability does not state that there is a causal relationship between the
two events, also it does not indicates that both events occurred simultaneously.
Conditional probability can be calculated as given below

P(X/Y) = P (X ∩ Y)/P(Y)

Where,
P(X/Y) – conditional probability or probability of event X occurring, given that event
Y has already taken place or occurred.
P (X ∩ Y) – Probability that both the events X and Y have already occurred.
P(Y) – Probability of event Y
Solution and steps:
In the mentioned case, team Garuda has the overall probability of winning the league
is 60%
Let us assume that
W- Garuda wins the league
L- Garuda lost the league
F1 – Garuda wins 1st game
F1’ – Garuda lost 1st game
Now taking the given data
P(W) = 0.60 P(L) = 1.0-0.60 = 0.40
P(F1/W) = 0.70 P(F1’/W) = 1.0-0.70 = 0.30
P(F1/L) = 0.25 P(F1’/L) = 1.0-0.25 = 0.75
Now let us develop a probability tree diagram for a better understanding
The probability tree diagram is used to represent the probability of events occurring
without the use of complicated formulas. It displays all the possible outcomes of an
event. A probability tree diagram can either represent a series of independent events
or can be used to denote conditional probabilities.
The probability tree diagram has two important parts namely
 Nodes
 Branches
The advantages of the tree diagram can be seen as discussed below
1. Simple visualization
2. Easy to construct
3. Easy to understand the dependencies
4. Good flexibility

In the above case, we will have a conditional tree diagram that is very useful in
depicting the outcome of the dependent events which means the effect of one event
can be seen on the other event.
From the above probability tree diagram, we can find that
Probability of Garuda winning 1st game and the series = P(W) X P(F1)
= 0.70 X 0.60
= 0.42
Probability of Garuda losing its 1st game and still winning the series = P(W) X P(F1’)
= 0.60 X 0.30
= 0.18
Probability of Garuda loosing 1srt game and loosing the series = P(F1’) X P(L)
= 0.75 X 0.40
= 0.30
Probability of Garuda winning the 1st game but loosing the series = P(F1) X P(L)
= 0.25 X 0.40
= 0.1
According to the case provided we need to find
Probability of Garuda losing the 1st game but still winning the series
P(W/F1’) = P (W ∩ F1’) / P(F1’)
= 0.18 / 0.18 + 0.30
= 0.18 / 0.48
= 0.375
Hence the probability of Garuda winning the series after losing its 1 st game has been
calculated as 37.5%.
Conclusion:
The above discussion shows that the importance of probability in business decision-
making is far ahead of regular statistics.
From the case above we can conclude that with the conditional probability usage of
the probability tree diagram we can have strategic action by combining the chances,
and choices of the decision, the concept of the probability tree aims to note every
possible outcome of any of the events. just it has helped to find the probability of
Garuda winning after losing its 1st game.
References:
1. NMIMS study material
2. [Link]/data/prob.
3. [Link]
4. [Link]

ANS-2
Introduction:
Businesses nowadays use various sets of statistical tools to understand the ups and
downs and the swinging nature of the market. This usage of tools helps the
management to take important business decisions under the various aspects of
trends.
Among the various no. of tools, one of them is known as the regression analysis.
regression analysis is the process of constructing a mathematical model that can be
used to predict or determine one variable by another variable or other variables.
Concept and application:
Regression analysis evaluates how strongly related the two elements are to help to
make stronger business plans, decisions, and forecasts. It can help to understand
the relationship between the variables that affect the sales of the budgeting goals.
The regression analysis has various applications in the fields such as finance,
banking, share markets, etc. There are six fundamental and independent variables
for linear regression analysis
1. The dependent and independent variables show a linear relationship between
the slope and the intercept
2. Independent variables are not random
3. Value of the residual variance is not random
4. Value of residual error remains constant
5. Value of the residual error is not correlated
6. Residual error values follow the normal distributions

Regression analysis evaluates the strength of the correlation between independent

and dependent variables.
A simple regression analysis evaluates the relationship between two variables while
multiple regression analysis assesses the correlation between a dependent variable
and more than the independent variable.
Businesses can use regression analysis to predict future sales, evaluate growth
opportunities, explain past occurrences and make strategic decisions. This
Regression Analysis can be helpful for a person in managing their data from a
business point of view. Regression evaluation has various types
1. Linear
2. Square
3. Consecutive

Linear analysis is easy, multiple, and direct routine models while Non-Linear analysis
is mostly used for more complex data.

To determine the equation of the regression line we need to establish the equation
form. mathematically the slope-intercept form of the equation can be seen as
provided below

y = MX + b

Where m = slope
b = y-intercept of the line
x = value of an independent variable
slope can be found as

b = [ ∑ XY – (∑x) (∑y)/n]
[ ∑x² - (∑x)²/n]

However alternately,

SS (XY) = [ ∑ XY – (∑x) (∑y)/n]

SS (xx) = [ ∑x² - (∑x)² /n]
b = SS (XY) / SS (xx)

solutions and steps:

in the above-mentioned case, we need to calculate the line of regression for y on x
X Y X*Y X² Y²
5 5 25 25 25
5 5 25 25 25
4 5 20 16 25
4 5 20 16 25
5 5 25 25 25
5 5 25 25 25
3 5 15 9 25
5 5 25 25 25
4 4 16 16 16
4 4 16 16 16
4 4 16 16 16
4 4 16 16 16
3 4 12 9 16
3 4 12 9 16
4 4 16 16 16
2 4 8 4 16
3 3 9 9 9
3 3 9 9 9
3 3 9 9 9
2 3 6 4 9
2 2 4 4 4
2 2 4 4 4
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
3 1 3 9 1
1 1 1 1 1
∑X = 86 ∑Y = 93 ∑ X*Y =340 ∑ X² = 320 ∑ Y² = 377
From the table drafted above
SS (x*y) = ∑ X * Y – (∑ X) (∑Y)
N
= 340 – (86) (93)
27
= 340 - 7998
27
= 340 – 296.22
= 43.77
SS (x*x) = ∑ x² - (∑x²) / n
= 320 – (86)²/ 27
= 320 – 273.92
= 46.07
We know that slope = SS (x * y) / SS (x* x)
= 43.77/46.07
b(1) = 0.95
b(0) = ∑y / n – b1∑x²/n
= 93/27 – 0.95 x 86/27
= 3.44-3.025
= 0.419
The intercept form of the equation can be seen below
y = b(0) + b1*x
y = 0.419 + 0.95*x
now we have found the line of regression y on x
since the value of the slope (b1) has been calculated as 0.95, this means that for
every unit of increase in the value of x, the value of y is expected to increase by 0.95
times with all other parameters remaining constant.
the value of the regression equation is equal to or more than 0.9502, it gives out a
good link between the factors. Hence based on the above-provided case the
business level of Shivangi is growing up and high.

Conclusion:
From the above cases, we tend to understand that regression analysis evaluates
how strongly related the two elements which help in making the business plans,
decisions, and forecasts stronger and also helps to understand the relationship
between the variables.
References:
1. NMIMS study sources
2. Buyju’[Link]

ANS-3(a)
Introduction:
In the case of the decision-making frequency can be referred to as the meaning “how
likely an event is if an experiment is repeated many times”. Frequency in
mathematics is a measure of how often a quantity is present and represents the
chances of occurrences of that quantity. in other words, frequency depicts how many
times a particular quantity has occurred.
Concept & application:
Relative frequency is the portion of total frequency that is in any given class interval
in a frequency distribution.
It is the extension of frequency where each particular frequency is represented
relative to all the particular frequencies of different quantities. Relative frequencies
help us to place an event into a larger context, for example, a survey shows that 30
students in a class like physics session over the others but it does not shows
whether this pool of 30 students is the bigger group or the smaller group, but if we
knew that 30 out of 40 students (75%) of the students were in the favor of physics
then we have a higher grade value.
Also, relative frequency allows us to compare the values between the studies. for the
same example, if different schools were surveyed then they will have different output
values. The reference value of 30 students might be a higher value for one school
while being a lower value for another school depending on the size of the class.
Formulae:
To calculate the Relative Frequency we need to have two pieces of information
1. Count the event for a category
2. Total no. of events

Then
Relative frequency = count of events for a category/total no. of events

As already discussed, relative frequency distribution describes the relative

frequencies for all possible outcomes in any case. The distribution displays
percentages for all the possible results. These results can be presented in
distributions using a table or bar chart.

In the above-provided case,

Let us find the relative frequency by drafting the table for

References:
1. NMIMS study course
2. [Link]

ANS-3(b)
Introduction:
Collection, presentation, analysis, organization, and data interpretation and
observations are known as statistics. These data can be represented in various
methods like tables, pie charts, polygons, bar graphs, etc.
Concept & application:
Bar graphs are widely used as qualitative techniques for data grouping. These are
the pictorial representation of the data in the form of vertical or horizontal rectangular
bars. These are known as bar charts.
A bar chart is a statistical approach to representing the data through horizontal and
vertical rectangular bars. The length of each bar is proportional to the data being
represented.
Major characteristics of the bar charts can be discussed as follows
1. he bar charts are used to compare the different data among different groups.
2. Bar charts show the relationship with the help of two axes. One of the axes
shows the categories while the other axis represents the values.
Some of the properties of the bar charts can be seen as follows
 All bars to have a common base.
 Each column in the bar graph has to have equal width.
 Height of the bar should correspond to the data value.
 Distance between each bar should be the same.

Some of the advantages of the bar charts can be seen below

1. Summarizes the large set of data in a simple form.
2. Displays each category.
3. Clarifies the trend of data better than the table.
4. Helps in estimating the key values at a glance.

As provided in the case of the districts we can make a draft of a bar chart for
the data provided below
 1. From the above-drafted bar chart we can see that the top 2 districts with
MSMEs will be
 SPSR NELLORE (54059)
 PRAKASAM (45171)

In the case of a set of ungrouped data, the data sets those are describing the central
values are known as the central tendencies. The central tendency is defined as the
statistical measure that can be used to represent the entire distribution or a data set
using a single value called a measure of central tendency.
Measures of the central tendencies can be measured in 3 ways – mean, median,
and mode.
Mean- the sum of observations divided by the total number of observations.
Median- the middle or central value is an ordered set.
Mode – the most frequently occurring value in a data set.

The Median is one of the measures of central tendencies, it is the value of the given
data set that is the middle most observation, obtained by arranging the data in
ascending order is called the “Median” of the data.

The Median can be understood as the middle value of the data set.
The Median can be calculated by the given formulas below

Median for odd no. of events: = (n+1)th term /2

Median for even no. of events = [(n/2)th term + (n+1/2)th term] / 2

However, it is to be understood that to find the median, the data should be arranged
either in an ascending or in descending manner.

2. As in the provided case

Let us first arrange the data in an ascending manner before we calculate the median
value for the case

STATE DISTRICT MSME

ANDHRA PRADESH SRIKAKULAM 10895
ANDHRA PRADESH KURNOOL 15362
ANDHRA PRADESH ANANTHAPUR 21193
ANDHRA PRADESH KRISHNA 23231
ANDHRA PRADESH GUNTUR 25479
ANDHRA PRADESH EAST GODAVARI 26546
ANDHRA PRADESH CHITOOR 27670
ANDHRA PRADESH VISHAKHAPATNAM 29070
ANDHRA PRADESH VIZIANAGRAM 30186
ANDHRA PRADESH WEST GODAVARI 33541
ANDHRA PRADESH Y. S. R 37500
ANDHRA PRADESH PRAKASAM 45171
ANDHRA PRADESH SPSR NELLORE 54059

Solution & Steps:

Step-1:
We can find that there is data from 13 districts provided in the case.
Step-2:
We know that for an odd no. of data observations which is 13 in this case,
The median can be calculated by (n+1)/2th term
Hence
Median = (13+1)/2
= 7th term
Hence the 7th term in the table can be seen as the median value of the districts which
is CHITTOOR with 27670 no. of MSMEs
Now let us see the districts above the median value in the same ascending order
STATE DISTRICT MSME
VISHAKHAPATNA
ANDHRA PRADESH M 29070
ANDHRA PRADESH VIZIANAGRAM 30186
ANDHRA PRADESH WEST GODAVARI 33541
ANDHRA PRADESH Y.S.R 37500
ANDHRA PRADESH PRAKASAM 45171
ANDHRA PRADESH SPSR NELLORE 54059

References:
1. NMIMS study material
2. Byju’[Link]
3. [Link]