Professional Documents
Culture Documents
Data
Insufficiency Previous track record of individuals is needed to analyse spending
and repayment patterns.
Total count - 100
Gender Ethnicity
Male
Caucasian Asian African American
48%
African Amer-
ican
25%
Caucasian
Analysis of Fe-
male
50%
Population
Asian
52% Male Female
26%
Frequency
50
40 32 31
30
20
10 2
0
30 40 50 60 70 80 90 100
Age
Ethnicity vs credit rating
100%
90%
80%
46
50 49
54
Effect of
70%
Ethnicity?
60%
50%
24
40% 24
27
24
30%
20%
30
27
23 22
10%
0%
Q1 Q2 Q3 Q4
Business
On Marital Status
On Student or Working professionals
Questions On Age and Gender
2. Identifying favourable and non-favourable customers based on
complete credit utilization (Zero balance)
3. Category of people who can be targeted to maximize credit
limit usage, based :
On percentage of credit utilization
On Monthly income
Percentage of Credit Utilization
Who to target to 140
131
No. of People
80
customers have a utilization of
67 68
less than 85%.
60
This can be improved by
offering incentives like
cashbacks. 40
310
utilization 40
35
High income – Zero credit balance
implies low risk and good credit 30
Frequency
utilization 25
21
20
15
15
Monthly Income
Q1 3142.66 10
Q2 4776.64 6
Q3 6137.70 5
2
IQR = Q3-Q1 2995.04 1
0
1.5 * IQR 4492.55 3000 6000 9000 12000 15000 18000
Q1 - 1.5IQR -1349.90
Q3 + 1.5IQR 10630.25
For Population in the
category of Zero Credit
Balance, Spend per
card can be analysed
Spend per Card
Q1 500.25
Q2 744.75
Q3 1168
IQR = Q3-Q1 667.75
1.5 * IQR 1001.63
Q1 - 1.5IQR -501.38
Q3 + 1.5IQR 2169.63
Data Analysis
% of credit utilization Monthly Income analysis Monthly Income analysis (Zero
analysis (total population) Credit Balance) Age of Population
55.66
Mean 91.04 Mean 7958.5 Mean 5073 Mean 75
0.862
Standard Error 0.34 Standard Error 310.1 Standard Error 285 Standard Error 49
Median 90.13 Median 5828.3 Median 4777 Median 56
Standard
Mode 100.00 Mode 4187.6 Deviation 2703 Mode 66
Standard Standard Standard 17.24
Deviation 6.76 Deviation 6203.0 Sample Variance 7303624 Deviation 981
38477110 297.5
Sample Variance 45.67 Sample Variance .7 Kurtosis 4 Sample Variance 558
-
1.065
Kurtosis -0.87 Kurtosis 2.9 Skewness 2 Kurtosis 58
0.011
Skewness -0.13 Skewness 1.7 Range 14388 Skewness 496
Range 27.69 Range 31025.3 Minimum 1824 Range 75
Minimum 72.31 Minimum 1822.3 Maximum 16212 Minimum 23
Maximum 100.00 Maximum 32847.6 Sum 456566 Maximum 98
36416.7 3183409.
Sum 6 Sum 5 Count 90 Sum 22267
Count 400.00 Count 400.0 Count 400
Descriptive Analysis
• The sample consists of 100 people – 48% male and 52% female population. Half of the
population belongs to Caucasian and other half is divided equally between African American and
Asian.
• The bar graph represents the customers from different regions in each quartile and their credit
ratings. We could infer that in each region the customers are distributed approximately equally
into all quartiles.
• The box plot addresses the group of people with zero credit balance and ranges of the income.
For people in the higher income range, it becomes easier to pay off the debt but customers in
the lower income range are less favorable for credit worthiness as they might not be able to pay
off the debt.
• This box plot shows that the outliers on the top of the graph are outliers and the most favorable
customers.
• The graph of the population with zero credit balance is left skewed which indicates that the
number of non-favourable customers are higher than favourable customers.
Descriptive Analysis
• 91 people have credit utilisation below 85%.
• From the analysis based on credit utilisation the customers with a potential to spend
more can be identified and they can be given incentives, discounts and cash backs.
• The sample consisted of 90 people with the zero balance or 100% credit utilization. Out
of these 90 individuals, people with higher income are the favorable set of customers
whereas the people with lower income pose a risk of default.
Thank You .