MACROECONOMICS 1

ASSIGNMENT 2: MACROECONOMICS DATA ANALYSIS

Course Business statistics 1

Course code ECON1193

Lecturer’s name Thai Trung Hieu

Student name Nguyen Tuan Hung- S3934835

Student ID S3934835

Wordcount 1235
 A. DESCRIPTIVE TACTICS
1) Measure of central tendency and dispersion
a. Measure of central tendency
Not Churn Churn
Age Tenure Age Tenure
Mean 36.5 33.15 45.33 34.67
Mode 32 29 59 #N/A
Median 33.5 33.5 52 36

Table 1 : Central tendency of not churned and churned

According to Table 1, there is no mode in the tenure category in the CHURN category since
the tenure column in these datasets did not have frequently repeated values. As a result, the mode
is an inappropriate indicator of central tendency. Additionally, since there are no values below
the lower bounce (<Q1-1.5*IQR) and above the upper bounce (>Q3+1.5*IQR), there are no
outliers in age or tenure in both categories. Because it takes into account every value in the
datasets, the mean is the most appropriate measure for central tendency. The median, which only
represents the middle value for organized data sets, is inappropriate in this situation. The chart
demonstrates that there was a significant difference in age between customers who left and those
who stayed. The average age in the churned group is 45.33 years old, which is older than the
average age in the non-churned category (36.5 years old).
b. Measures of Variation

Not Churn Churn

Age Tenure Age Tenure
Standard Deviation 11.61 19.32 15.61 16.12
Sample Variance 134.79 373.08 243.81 259.81
IQR 17.25 40.75 26.5 25
Coefficient of Variance 0.32 0.58 0.34 0.46
Range 36 53 42 50

Table 2 : Measures of Variation of not churned and churned

Since this dataset had no outliers, according to section a, it is also suitable to quantify
using range, standard deviation, and coefficient of variation in addition to interquartile
range. More noteworthy is that the median of age and tenure in both categories are close
to its mean. Therefore, standard deviation, which uses all of the original units of data,
should be considered the best measure of variation in this situation.
2) Histograms and Box plot
a) Histograms

8
7
7

6
5
5
uency

4 4
4
 2

0
18-27 28-37 38-47 48-57
7
6
6

Frequency
4
3 3
3
2
2
1
1

0
18-27 28-37 38-47 48-57 58-67

Figure 1 : Age of non-churning customers Figure 2 : Age of churning customers

The two images up above show the age histogram between non-churning and churning
consumers. In terms of non-churning clients, the distribution is right-skewed, with the mean
value (which is 36.5 years old) being higher than the median value (33.5 years old). Because of
this, the histogram's values are more concentrated on the left side of the graph, as seen by the
right side's lengthy tail. In contrast, the age of churning customers' histograms have a left-skewed
shape distribution and a mean value that is lower than the median (45.33<52 years old), which
means that more values in these histograms are concentrated on the right side of the graph, as
seen by the left-side graph's long tail. The highest frequency of the category of customers who do
not leave is between the ages of 28-37, while the highest frequency of customers leaving is 58-67
in the age group.

3.5
3 3 3
7 3
6
6 2.5
5 2 2 2
Frequency

4 2
Frequency

4
3 3 1.5
3
2 2
2 1
1 0.5
0
6-15 16-25 26-35 36-45 46-55 56-65 0
6-15 16-25 26-35 36-45 46-55 56-65

Figure 3 : Tenure of non-churning clients Figure 4 : Tenure of churning clients

 Using two figures above, With smaller mean values than median, 33.15–33.5 months for
customers who haven't churned, and 34.67–36 months for those who have, both datasets'
distribution shapes are left-skewed. As a result, more values in these two histograms are
concentrated on the right side of the graph, as shown by a long tail on the left.
b) Box and whisker plot

Figure 5 : Age of non-churning and churning

In terms of the non-churning, it was a positive skewed shape, since Q3-Median >
Median-Q1(13.5>4.25). While the churn was a negative skewed shape because Q3-
Median < Median-Q1(7<22).

Figure 6 : Tenure of non-chunrning and churning

 In this case, both tenure of non-churning and churning are positive skewed
shape because Q3 - Median < Median - Q1, 19.25 < 23 for not churned and 12 < 15 for
churned category.

B. Probability
1.
Proportion = (15/35)*100 = 42.85%

The ratio is calculated by taking the total number of customers who have churned are
15 people divided by the total number of customers in the data sets ( 35 people ) and then
multiplying by 100. The result is that the churn rate accounts for 42.85% of the
company's total customers and the number of consumers who remain with the firm
accounts for 57.15%.

2. Assume event C : customers whose tenure above 2 years

P(C) = 23/35
P(A∩C) = 10/35
P(AlC) = P(A∩C)/P(C) = (10/35)/(23/35) = 0.(4347)

To determine the relationship between tenure and churn, we must first calculate P(A)
with the event A being customers churn inferring P(A)= 15/35 = 0.4285. The results show
that P(A) and P(AlC) are not equal, showing that churn is affected by tenure, so there is a
relationship between churned and tenure.

3. Event D : customers whose age less than 30

P(D) = 8/35
P(A∩D)=3/35
P(AlD) = P(A∩D)/P(D) = 0.375

It can be seen that the number of customers leaving under 30 years old accounts for
37.5% of total customer churn. Therefore, it can be concluded that customers who are
less than 30 years old are not likely to churn.

C. Discussion and Recommendation

1. Customer retention and loyalty is becoming a crucial factor in the success of the
telecom providers in today's cutthroat industry( Shafei, Ingy et al 2016). Service
quality was defined as the overall assessment of a particular service company that
comes from contrasting that business's performance with the general expectations of
consumers for how companies in that sector should operate( Shafei, Ingy et al 2016).
The second is brand image, Malik stated that brand image is a circumlocutory
instrument that may influence customer behavior in a good way and significantly
enhance any company performance(Bina Nazir et al 2016). Khan (2012) reported in
 his research that brand awareness, brand affinity, and brand familiarity had a big
influence on customer retention(Bina Nazir et al 2016).
2. There are 2 factors that have a higher probability of leaving. The first is the elderly
segment, as shown in Figure 2, older people aged 58-67 have a higher propensity to
leave. Second, it is the probability of customers leaving with a shelf life of more than
2 years, with 43.47% of total customers leaving, which can be seen as a trait
associated with a higher probability of leaving.

3. The author will utilize Mobi Fone evidence to assess individual analysis. Experts
claim that MobiFone has been successful in keeping consumers up to this point
because to its culture of customer care, particularly its capacity to comprehend and
continually absorb user comments to enhance goods and services (Tram N 2022).
Additionally, Mobifone has a strong reputation for quality and customer service,
which serve as the company's distinctive symbols. As a consequence, MobiFone
received votes for "The mobile network with the best service and customer care" and
"The most popular mobile network" for a number of years in a row (2005-2010)
(Tram N 2022). It proves that what the author analyzes in the above sections is valid.

4. Many options assist telecommunications companies in lowering the number of clients

quitting. Start by finding ways to expand reward programs for members with tenures
of more than two years in order to attract new clients and boost the happiness of
existing ones. Secondly, it is more interested in customer service, hence, the elderly
are often quite tough, so careful treatment may make them feel more content and
persist with the service of the telecommunications firm.

II. Reference

1. Trâm, N. (2022) Mobifone - Hành Trình 29 năm Tiên Phong Trong Chất Lượng Chăm
Sóc Khách Hàng, Báo Kinh tế đô thị. Available at: https://kinhtedothi.vn/mobifone-hanh-
trinh-29-nam-tien-phong-trong-chat-luong-cham-soc-khach-hang.html (Accessed: 24
July 2023).
2. Bina, N., Muhammad, A. and Mahwish, J. (2016) The impact of brand image on the
customer retention: A mediating role ..., The Impact of Brand Image on the Customer
Retention: A Mediating Role of Customer Satisfaction in Pakistan. Available at:
https://www.researchgate.net/profile/Mahwish-Jamil-2/publication/338775113_The_Imp
act_of_Brand_Image_on_the_Customer_Retention_A_Mediating_Role_of_Customer_Sa
tisfaction_in_Pakistan/links/62fe4e20aa4b1206fabd0ee6/The-Impact-of-Brand-Image-
on-the-Customer-Retention-A-Mediating-Role-of-Customer-Satisfaction-in-Pakistan.pdf
(Accessed: 24 July 2023).
3. Shafei, I. and Tabaa, H. (2016) ‘Factors affecting customer loyalty for mobile
telecommunication industry’, EuroMed Journal of Business, 11(3), pp. 347–361.
doi:10.1108/emjb-07-2015-0034.