compat=1.

18
Bongu Ashish(23BDS014)
Ishan Srivastava(23BDS023) Latex Assignment 1
Taran Jain(23BDS062) https://www.overleaf.com/read/hnhznbgjtzdsc96585(Read- 2024-09-15
Shaik Izhaar Ahmed(23BDS053)
Only link to the project)
Shivansh Shukla(23BDS054)

1 First Chapter

CASE STUDY : ENTERTAINMENT ON A CRUISE SHIP(Problem on Chapter 1 Page 16)

1.1 First Subtask

1.1.1 Problem Description

The problem is to match the entertainment offerings with the preferences of passengers while
considering budget, venue, and contractual constraints. The aim is to maximize passenger
satisfaction by providing entertainment that caters to a wide range of tastes.

1.2 Second Subtask

Solution Approach

To solve this, we can gather real-time data on passenger preferences and attendance, then
adjust the entertainment schedule dynamically to maximize satisfaction while adhering to
the constraints.

1.3 Third Task

1.3.1 Variables

• Let P = {p1 , p2 , . . . , pn } be the set of passengers,

• Let E = {e1 , e2 , . . . , em } be the set of entertainment options,
• Let Aij be the attendance for entertainment option ej by passenger pi ,
• Let Dij be the demand (or preference) for entertainment option ej by passenger pi .

Fourth Task
1.3.2 Objective

The goal is to maximize the total satisfaction of passengers, defined as:

n X
X m
Maximize S = Dij Aij
i=1 j=1

Fifth Subtask
1.3.3 Constraints

• Budget constraint: The total cost of entertainment must not exceed the budget B:
Xm
Cj ≤ B
j=1

where Cj is the cost of entertainment ej .

• Venue constraint: The total venue space used by the entertainment must not exceed
1
the available space Vmax :
m
X
Vj ≤ Vmax
2 Second Chapter

Analysis of Dow Jones Industrial Average Percentage Changes:(Problem on Chapter 2 page

30 Level B 5 th problem)

2.1 Introduction
This document outlines the process for recoding monthly percentage changes of the Dow
Jones Industrial Average (DJIA) and visualizing the distribution of these changes.

2.2 Recoding Percentage Changes

We categorized the percentage changes into the following six categories:
• Large negative: < −3%
• Medium negative: −3% ≤ Change < −1%
• Small negative: −1% ≤ Change < 0%
• Small positive: 0% ≤ Change < 1%
• Medium positive: 1% ≤ Change < 3%
• Large positive: ≥ 3%
In Excel, this recoding can be accomplished with the following formula:
=IF(B2 < -3%, "Large negative",
IF(B2 < -1%, "Medium negative",
IF(B2 < 0%, "Small negative",
IF(B2 < 1%, "Small positive",
IF(B2 < 3%, "Medium positive",
"Large positive")))))

2.3 Column Chart

A column chart was created to display the counts of each category. Below is an example of
how to create such a chart using LaTeX.
30
30
25
20
20
Count

15
10
10
5

0
e

ve
e

ve

ve

Category
tiv

tiv

iv

iti
iti

iti
at
ga

ga

s
os

s
eg

po
po
ne

ne

lp
ln

e
um
al
e

rg
al
rg

Sm
iu

La
Sm

i
La

ed
ed

M
M

2
2.4 Commentary
The shape of the column chart reveals the distribution of percentage changes. Observing the
chart, we can identify whether changes are more frequently small or large, and whether
there is a predominant trend in the data.

3 Third Chapter: Problem statement on

3.1 Introduction
This document demonstrates how to use pivot tables to drill down through hierarchical data
and visualize the results using LaTeX. We use the ‘Supermarket Transactions.xlsx‘ file as an
example.

3.2 Hierarchical Data

Consider the following hierarchy for products:
• Product Family
• Product Department
• Product Category
• Product Type
We will drill down from the broadest category, such as ‘Drink‘, to more specific subcategories
like ‘Beer and Wine‘.

3.3 Pivot Table Analysis

To analyze revenue data, follow these steps:
1. Create a Pivot Table in Excel with ‘Product Family‘ as the row label and ‘Revenue‘ as the
values.
2. Drill down to see detailed revenues by ‘Product Department‘, ‘Product Category‘, and
‘Product Type‘.
3. Similarly, analyze revenues by location and time hierarchies.

3.4 Visualization Using pgfplots

The following plots illustrate hierarchical revenue data with increased bar width and size:

3
3.4.1 Revenue by Product Family

Revenue by Product Family

·104
50,000
5

30,000
Revenue

20,000
2

0
k

ck

ld
Product Family
in

ho
a
Dr

Sn

se
ou
H
3.4.2 Revenue by Product Department (Drilled Down from Drink)

Revenue by Product Department

·104
30,000
3

2.5
20,000
2
Revenue

1.5

0.5

0
s

es

Product Department
ge

ag
ra

r
ve

ve
Be

Be
ic

lic
ol

o
oh

oh
c

lc
Al

-A
on
N

4
3.4.3 Revenue by Product Category (Drilled Down from Alcoholic Beverages)

Revenue by Product Category

·104
1.6 15,000 15,000
1.4
1.2
1
Revenue

0.8
0.6
0.4
0.2
0
er

e
Product Category

in
Be

W
3.4.4 Revenue by Product Type (Drilled Down from Beer)

Revenue by Product Type

8,000
8,000
7,000

6,000
Revenue

4,000

2,000

0
r
A

Product Type
ge
IP

La

3.5 Conclusion
By drilling down through hierarchical data, managers can gain detailed insights into reve-
nues. The provided plots illustrate the revenue distribution at different levels of the product
hierarchy with enhanced visual appeal.

5
4 Fourth Chapter
4.1 Yankees vs Phillies Best-of-Three and Best-of-Five Series (Page 164 Problem 37

Problem Overview
We are given the following:
• The Yankees have a probability of winning at home PY = 0.55,
• The Phillies have a probability of winning at home PP = 0.53,
• The games are independent of each other,
• The series is a best-of-three (parts a, b, and c) and best-of-five (part d).

Part a: Probability that the Yankees win the series (Best-of-Three)

For the best-of-three series, the series ends as soon as one team wins two games. The series
can end in two games or three games. The different ways the Yankees can win the series are
as follows:
• Yankees win both first two games: This occurs with probability:

P (YY) = PY × (1 − PP ) = 0.55 × 0.47 = 0.2585

• Yankees win in three games (lose the second game but win first and third): This
occurs with probability:
P (YNY) = PY × PP × PY = 0.55 × 0.53 × 0.55 = 0.1604

• Yankees win in three games (lose the first game but win second and third): This
occurs with probability:
P (NYY) = (1 − PY ) × PP × PY = 0.45 × 0.53 × 0.55 = 0.1312

Thus, the total probability that the Yankees win the series is:
P (Yankees win) = P (YY) + P (YNY) + P (NYY) = 0.2585 + 0.1604 + 0.1312 = 0.5501

Part b: Distribution of Net Winnings and Mean/Standard Deviation

Let W be the random variable representing your net winnings in a single game:
• If the Yankees win a game, you win $100,
• If the Yankees lose a game, you lose $105.
The probabilities of each outcome are as follows:
• P (Win) = PY ,
• P (Lose) = 1 − PY .

6
 Yankees’ Chances in Best-of-Three Series
0.3
0.26

0.2

Probability
0.16
0.13

0.1

0
YY YNY NYY
Outcome

Abbildung 1: Probability of Yankees winning the series in different ways.

Thus, the probability mass function (PMF) for your net winnings is:
P (W = 100) = 0.55, P (W = −105) = 0.45
The expected value of your net winnings is:
E[W ] = 100 × 0.55 + (−105) × 0.45 = 55 − 47.25 = 7.75
The variance of W is:
Var(W ) = (100 − 7.75)2 × 0.55 + (−105 − 7.75)2 × 0.45
Var(W ) = (92.25)2 × 0.55 + (−112.75)2 × 0.45
Var(W ) = 8462.06 × 0.55 + 12713.56 × 0.45 = 4654.13 + 5721.10 = 10375.23
Thus, the standard deviation is:
√
σW = 10375.23 ≈ 101.86
The expected value ($7.75) indicates this betting strategy is favorable to you.

Part c: Probability with Games in Philadelphia, New York, Philadelphia

(Best-of-Three)
Now, the series is played in the following order: Philadelphia, New York, Philadelphia. The
different ways the Yankees can win the series are:
• Yankees win both first two games:
P (YY) = (1 − PP ) × PY = 0.47 × 0.55 = 0.2585
• Yankees win in three games (lose the second game but win first and third):
P (YNY) = (1 − PP ) × PY × (1 − PP ) = 0.47 × 0.55 × 0.47 = 0.1218
• Yankees win in three games (lose the first game but win second and third):
P (NYY) = PP × PY × (1 − PP ) = 0.53 × 0.55 × 0.47 = 0.1373
Thus, the total probability that the Yankees win the series is:
P (Yankees win) = 0.2585 + 0.1218 + 0.1373 = 0.5176
This shows the ”home field advantageßlightly helps the Phillies.

7
 Distribution of Net Winnings
0.6
0.55

0.45

0.4

Probability
0.2

0
100 -105
Net Winnings

Abbildung 2: Distribution of Net Winnings from betting on Yankees games.

Part d: Probability that the Yankees win a Best-of-Five Series

In the best-of-five series, the Yankees need to win three games. The games alternate between
New York and Philadelphia, starting in New York.

5 Chapter 5 (Problem statement 76 on page 218

6 Analysis of Credit Write-Offs

6.1 (a) Probability of Default and Write-Off Greater Than $250
To find the probability that a customer defaults and the write-off is greater than $250:
Write-off = 0.80 · X

We need:
250
0.80 · X > 250 =⇒ X > = 312.50
0.80
We standardize X:
X −µ 312.50 − 350
Z=
= = −0.375
σ 100
The probability of defaulting and having a write-off greater than $250 is:

P (Write-off > 250) = p · P (X > 312.50) = 0.07 · (1 − Φ(−0.375))

6.2 (b) Mean and Standard Deviation of Customers with Write-Off Greater Than $250
Let N = 500 be the number of customers.
Let pwrite-off be the probability of write-off greater than $250. This is:
pwrite-off = P (X > 312.50) = 1 − Φ(−0.375)

The number of customers meeting this criterion follows a binomial distribution:

Y ∼ Binomial(500, pwrite-off )

8
 Normal Distribution of Credit Charges
·10−2
4
Normal Distribution
Area for X > 312.50

3
Probability Density

0
200 250 300 350 400 450 500
Credit Charges ($)

Abbildung 3: Normal Distribution of Credit Charges with Highlighted Area for X > 312.50.

Mean and standard deviation for this binomial distribution are:

Mean = N · pwrite-off
p
Standard Deviation = N · pwrite-off · (1 − pwrite-off )

Approximate Normal Distribution for Binomial

·10−2
Normal Approximation

6
Probability

−5 0 5 10 15 20 25 30 35 40 45 50 55
Number of Customers with Write-Off ¿ $250

Abbildung 4: Normal Approximation for the Number of Customers with Write-Off Greater Than $250.

6.3 (c) Probability That At Least 25 Customers Meet the Description

Using the normal approximation:
Y ≈ Normal (Mean, Variance)

9
We calculate:  
25 − Mean
P (Y ≥ 25) ≈ 1 − Φ
Standard Deviation

Probability of At Least 25 Customers

·10−2
Normal Distribution
Area for Y ≥ 25
6
Probability

−5 0 5 10 15 20 25 30 35 40 45 50 55
Number of Customers with Write-Off ¿ $250

Abbildung 5: Normal Approximation for At Least 25 Customers with Write-Off Greater Than $250.

6.4 (d) Simulation of Bad Debt from 500 Customers with No Recovery
To simulate the total amount of bad debt with no recovery:
• **Binomial Calculation**: Calculate the total amount based on the number of defaults.
• **Normal Calculation**: Approximate the total amount using the normal distribution.
Binomial Calculation:
Total Bad Debt = Number of Defaults × 350
Normal Calculation:
p
Total Bad Debt = Normal(500 × 0.07 × 350, 500 × 0.07 × (1 − 0.07) × 350)

10