Department of Business Administration: Final Exam (Take Home) Summer - 2020

Final Exam (Take Home) Summer - 2020
Subject: Business Analytics Max. Marks: 40
Instructor: Dr. Noman Islam/ Muizz Alam Khan Program: BBA

Monis Ahmed Thakur
Department of Business Administration
Instructions:
1. Attempt all questions manually on paper and then submit it as a single pdf document. This can
be achieved by taking clear pictures of the sheets and then pasting these on word document.
(You can use applications like Microsoft Lens etc. for this purpose to create a scanned
document). Please write page number, name and/or reg. no on each page.
2. The document name should be of the format yournameregistrationnumber.pdf. For example,
AhmadKhan21345.pdf
3. Write your name and registration ID on the first page of your Word file.
4. Answer scripts can be uploaded on LMS any time before its deadline. Therefore, do not wait for
the last hour to avoid any unforeseen problems.
5. Submission of answer copy(ies) will be considered acceptable through LMS only. Therefore, do
not submit your document through email or any other medium.
6. Please do not use pencils except for underlining or drawing diagrams.
7. Marks of each question are mentioned at the end of the question.
8. Any attempt to use unfair means will disqualify you from the examination.
9. Please ensure you are uploading the correct document.
Page 1 of 9
Q1
The Codes have been written in R and are attached at the end of the answer for reproduction.
a)
Alpha has a mean of 13 while Bravo has a mean of 11. Here are the boxplots
From the box-plot is it evident that Alpha serving point is better than Bravo in mean, median(12.5 to
10) and Ranges. So, Alpha serving point is much more efficient than Bravo
b)
The sd for Alpha is 3.92 while the mean for Bravo is 5.46
The z-score is given by (18-mean)/sd
So, the z-score for Alpha is 1.276885 and the z-score for Bravo is 1.287593
Since the z-score of Alpha has a lower absolute value, it is handling customer better
The R codes are-
Alpha=c(12,16,11,10,17,10,13,18,6,17)
Bravo=c(10,3,8,7,6,16,15,16,20,9)
boxplot(Alpha)
Page 2 of 9
boxplot(Alpha, Bravo,
main = "Boxplots of Alpha and Bravo",
names = c("Alpha", "Bravo"),
las = 2,
col = c("orange","yellow"),
border = "brown"
)
mean(Alpha)
mean(Bravo)
sd(Alpha)
sd(Bravo)
z_alpha= (18-mean(Alpha))/sd(Alpha)
z_bravo= (18-mean(Bravo))/sd(Bravo)
Page 3 of 9
Question 2:
a.
Page 4 of 9
Question 3:
Let be the population mean.
a) At a two-tailed test with α = 0.05
vs
b. At a left-tailed test with α = 0.15
vs
Please give thumbs up if you like the answer
Question 4: Consider the following transactions data of a company: [06 marks]

a , b , c , d , e, f
a , f, c , d
b, c, d, e
e, f , a , d
e, f , g , h , i
Find the confidence and lift ratio for following rules:
a. a, c → f
b. d, f → g
c. b, c → e
d. f → b
e. e, f → g
f. h→ i
Page 5 of 9
Question 5: Suppose your class of Marketing got the following results in Summer 2020 (one
record for each student): [05 marks]
Table 3: Result of Marketing Students
Student Id Communication Skills Punctuality Class Participation Final
1 15 4.4 1.4 4.2
2 13 2.3 2.2 2.4

3 4.5 1.9 1.1 6.1
4 5.2 1.4 1.3 8.2

5 6.7 7.9 1.5 2.2
6 7.1 5.6 3.6 9.9

7 1.2 1.3 4.8 2.4
8 1.1 1.2 9.9 1.8
a. Using k-means clustering, divide the data into three groups. Show your calculations
up to two iterations. [03 marks]
b. What analysis do you make after performing this descriptive data mining step?
Discuss the basic characteristics of each of these two groups of students [02 marks]
Question 6:
4 pic send ki line se karta ja
Question 7: 1. Given data is saved in CSV file with name HCData.
2. Code to perform heirarchial clustering by single linkage method
setwd("D:/HC")
#Reading data
mydata = read.csv("HCData.csv",header = TRUE)
attach(mydata)
# Dissimilarity matrix
d <- dist(mydata, method = "euclidean")
# Hierarchical clustering using Complete Linkage

HC1 <- hclust(d, method = "single" )
Page 6 of 9
# Plot the obtained dendrogram
plot(HC1, cex = 0.6, hang = -1)
3. Dendrogram for single linkage method:
4. Code to perform heirarchial clustering by average linkage method
setwd("D:/HC")
#Reading data
attach(mydata)

HC1 <- hclust(d, method = "average" )

5. Dendrogram for average linkage method:
Page 7 of 9
6. Code to perform heirarchial clustering by complete linkage method
setwd("D:/HC")
#Reading data
attach(mydata)

HC1 <- hclust(d, method = "complete" )

7. Dendrogram for complete linkage method:
Page 8 of 9
8. In any case, for sure there will be atleast a small difference present in clustering outcome by using
different method. In our case, upto the height of 60, the outcome will not be different. Or in other
terms, considering the data as three or above clusters, our cluster outcome will be same for all the
three cases.
Page 9 of 9

Department of Business Administration: Final Exam (Take Home) Summer - 2020

Uploaded by

Copyright:

Available Formats

Department of Business Administration: Final Exam (Take Home) Summer - 2020

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Department of Business Administration: Final Exam (Take Home) Summer - 2020

Uploaded by

Copyright:

Available Formats

Final Exam (Take Home) Summer - 2020

Subject: Business Analytics Max. Marks: 40

Instructor: Dr. Noman Islam/ Muizz Alam Khan Program: BBA

Department of Business Administration

The z-score is given by (18-mean)/sd

The R codes are-

a) At a two-tailed test with α = 0.05

b. At a left-tailed test with α = 0.15

Please give thumbs up if you like the answer

Question 4: Consider the following transactions data of a company: [06 marks]

2 13 2.3 2.2 2.4

4 5.2 1.4 1.3 8.2

6 7.1 5.6 3.6 9.9

8 1.1 1.2 9.9 1.8

Question 7: 1. Given data is saved in CSV file with name HCData.

2. Code to perform heirarchial clustering by single linkage method

# Hierarchical clustering using Complete Linkage

3. Dendrogram for single linkage method:

4. Code to perform heirarchial clustering by average linkage method

# Hierarchical clustering using Complete Linkage

# Plot the obtained dendrogram

5. Dendrogram for average linkage method:

# Hierarchical clustering using Complete Linkage

# Plot the obtained dendrogram

7. Dendrogram for complete linkage method:

You might also like