library(readr)
## Warning: package 'readr' was built under R version 4.3.3
credit <- read_csv("C:/Users/hp/Desktop/dataset/Credit Card Customers/BankChurners.csv")
## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
## dat <- vroom(...)
## problems(dat)
## Rows: 10127 Columns: 21
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (7): Attrition_Flag, Gender, Education_Level, Marital_Status, Income_Ca...
## dbl (14): CLIENTNUM, Customer_Age, Dependent_count, Months_on_book, Total_Re...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
credit
## # A tibble: 10,127 × 21
## CLIENTNUM Attrition_Flag Customer_Age Gender Dependent_count Education_Level
## <dbl> <chr> <dbl> <chr> <dbl> <chr>
## 1 768805383 Existing Custo… 45 M 3 High School
## 2 818770008 Existing Custo… 49 F 5 Graduate
## 3 713982108 Existing Custo… 51 M 3 Graduate
## 4 769911858 Existing Custo… 40 F 4 High School
## 5 709106358 Existing Custo… 40 M 3 Uneducated
## 6 713061558 Existing Custo… 44 M 2 Graduate
## 7 810347208 Existing Custo… 51 M 4 Unknown
## 8 818906208 Existing Custo… 32 M 0 High School
## 9 710930508 Existing Custo… 37 M 3 Uneducated
## 10 719661558 Existing Custo… 48 M 2 Graduate
## # ℹ 10,117 more rows
## # ℹ 15 more variables: Marital_Status <chr>, Income_Category <chr>,
## # Card_Category <chr>, Months_on_book <dbl>, Total_Relationship_Count <dbl>,
## # Months_Inactive_12_mon <dbl>, Contacts_Count_12_mon <dbl>,
## # Credit_Limit <dbl>, Total_Revolving_Bal <dbl>, Avg_Open_To_Buy <dbl>,
## # Total_Amt_Chng_Q4_Q1 <dbl>, Total_Trans_Amt <dbl>, Total_Trans_Ct <dbl>,
## # Total_Ct_Chng_Q4_Q1 <dbl>, Avg_Utilization_Ratio <chr>
# Question 1 : What is the average monthly spending of customers on their credit cards?
average_monthly_spending <- mean(credit$Total_Trans_Amt)
cat("Average monthly spending of customers is : ",average_monthly_spending)
## Average monthly spending of customers is : 4404.086
# Question 2 : What is the highest credit limit among customers?
library('dplyr')
## Warning: package 'dplyr' was built under R version 4.3.3
##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
highest_credit_limit <- max(credit$Credit_Limit)
cat("The highest credit limit among customers is : ",highest_credit_limit)
## The highest credit limit among customers is : 34516
# Question 3 : What are the distinct values for marital status among customers?
library('dplyr')
distinct_marital_status <- credit %>%
distinct(Marital_Status)
distinct_marital_status
## # A tibble: 4 × 1
## Marital_Status
## <chr>
## 1 Married
## 2 Single
## 3 Unknown
## 4 Divorced
# Question 4 : How does the average monthly spending vary across different education levels?
library('dplyr')
avg_monthly_spending <- credit %>%
group_by(Education_Level) %>%
summarise(avg_monthly_spending = round(mean(Total_Trans_Amt),2))
avg_monthly_spending
## # A tibble: 7 × 2
## Education_Level avg_monthly_spending
## <chr> <dbl>
## 1 College 4256.
## 2 Doctorate 4193.
## 3 Graduate 4418.
## 4 High School 4404.
## 5 Post-Graduate 4515.
## 6 Uneducated 4484.
## 7 Unknown 4420.
# Question 5 : What is the customer attrition rate of credit card customers?
attrition_rate <- mean(credit$Attrition_Flag == "Attrited Customer")
cat("Attrition Rate of credit card customers is : ",attrition_rate)
## Attrition Rate of credit card customers is : 0.1606596
# Question 6 : Which credit card products are the most popular among customers?
card_category_frequency <- table(credit$Card_Category)
sorted_data <- sort(card_category_frequency, decreasing = TRUE)
sorted_data
##
## Blue Silver Gold Platinum
## 9436 555 116 20
# Question 7 : How many customers are there with a credit limit greater than $10,000?
library('dplyr')
customers_gt_10000 <- credit %>%
filter(Credit_Limit > 10000) %>% select(Credit_Limit, everything())
customers_gt_10000
## # A tibble: 2,754 × 21
## Credit_Limit CLIENTNUM Attrition_Flag Customer_Age Gender Dependent_count
## <dbl> <dbl> <chr> <dbl> <chr> <dbl>
## 1 12691 768805383 Existing Customer 45 M 3
## 2 34516 810347208 Existing Customer 51 M 4
## 3 29081 818906208 Existing Customer 32 M 0
## 4 22352 710930508 Existing Customer 37 M 3
## 5 11656 719661558 Existing Customer 48 M 2
## 6 11751 710599683 Existing Customer 56 M 1
## 7 30367 709967358 Existing Customer 48 M 4
## 8 13535 753327333 Existing Customer 41 M 3
## 9 14470 709327383 Existing Customer 45 F 2
## 10 20979 806165208 Existing Customer 47 M 1
## # ℹ 2,744 more rows
## # ℹ 15 more variables: Education_Level <chr>, Marital_Status <chr>,
## # Income_Category <chr>, Card_Category <chr>, Months_on_book <dbl>,
## # Total_Relationship_Count <dbl>, Months_Inactive_12_mon <dbl>,
## # Contacts_Count_12_mon <dbl>, Total_Revolving_Bal <dbl>,
## # Avg_Open_To_Buy <dbl>, Total_Amt_Chng_Q4_Q1 <dbl>, Total_Trans_Amt <dbl>,
## # Total_Trans_Ct <dbl>, Total_Ct_Chng_Q4_Q1 <dbl>, …
# Question 8 : Filter the data to include customers with a credit limit greater than $5000 and calculate the mean customer age for each income category.
library('dplyr')
filtered_data <- credit %>% filter(Credit_Limit > 5000) %>% group_by(Income_Category) %>% summarise(Mean_customer_age = round(mean(Customer_Age)))
filtered_data
## # A tibble: 6 × 2
## Income_Category Mean_customer_age
## <chr> <dbl>
## 1 $120K + 48
## 2 $40K - $60K 46
## 3 $60K - $80K 46
## 4 $80K - $120K 46
## 5 Less than $40K 46
## 6 Unknown 46
# Question 9 : What is the distribution of months of inactivity (Months_Inactive_12_mon) across different clients?
library('tidyr')
inactive_months <- credit %>%
select(CLIENTNUM, Months_Inactive_12_mon) %>%
gather(key = "Months Activity", value = "Value", -CLIENTNUM)
inactive_months
## # A tibble: 10,127 × 3
## CLIENTNUM `Months Activity` Value
## <dbl> <chr> <dbl>
## 1 768805383 Months_Inactive_12_mon 1
## 2 818770008 Months_Inactive_12_mon 1
## 3 713982108 Months_Inactive_12_mon 1
## 4 769911858 Months_Inactive_12_mon 4
## 5 709106358 Months_Inactive_12_mon 1
## 6 713061558 Months_Inactive_12_mon 1
## 7 810347208 Months_Inactive_12_mon 1
## 8 818906208 Months_Inactive_12_mon 2
## 9 710930508 Months_Inactive_12_mon 2
## 10 719661558 Months_Inactive_12_mon 3
## # ℹ 10,117 more rows
# Question 10 : What is the lifetime value of credit card customers?
# Calculate lifetime value for a one-year time frame
# Assuming revenue = total transaction amount
lifetime_value <- sum(credit$Total_Trans_Amt)
cat("Lifetime Value of credit card customers : ",lifetime_value)
## Lifetime Value of credit card customers : 44600182
# Question 11 : What is the distribution of customer ages in the dataset?
library('ggplot2')
ggplot(credit, aes(x = Customer_Age)) +
geom_histogram(binwidth = 5, fill = "lightgreen", color = "black") +
labs(title = "Distribution of Customer Ages",
x = "Age",
y = "Frequency")
# Question 12 : How does the length of time a customer(gender wise) has been with the bank relate to their credit card usage?
library('ggplot2')
ggplot(credit, aes(x = Months_on_book, y = Total_Trans_Amt, col = Gender)) +
geom_point() +
labs(title = "Relationship Between Length of Time as Customer and Credit Card Usage",
x = "Months on Book",
y = "Total Transaction Amount")
# Question 13 : What is the distribution of credit card customers by income level and gender?
library('ggplot2')
ggplot(credit, aes(x = Income_Category, fill = Gender)) +
geom_bar(position = "stack") +
labs(title = "Distribution of Credit Card Customers by Income Level and Gender",
x = "Income Level",
y = "Count") +
scale_fill_manual(labels = c("Female","Male"),
values = c("skyblue", "blue"))
# Question 14 : What is the distribution of customers across different income categories?
pie(table(credit$Income_Category),
main = "Distribution of customers across different income categories",
labels = unique(credit$Income_Category),
col = rainbow(length(unique(credit$Income_Category))))
# Question 15 : What is the distribution of education levels among individuals in the dataset?
library('ggplot2')
education_counts <- credit %>%
count(Education_Level) %>%
arrange(desc(n))
ggplot(education_counts, aes(x = reorder(Education_Level, n), y = n)) +
geom_bar(stat = "identity", fill = rainbow(length(unique(credit$Education_Level)))) +
labs(title = "Distribution of Education Levels",
x = "Education Level",
y = "Count")
# Question 16 : How does the total transaction amount influence the credit limits assigned to customers?
library('dplyr')
library('ggplot2')
credit_data <- sample_n(credit,1000) %>%
select(Credit_Limit, Customer_Age, Income_Category, Total_Trans_Amt)
ggplot(credit_data, aes(y = Credit_Limit, x = Total_Trans_Amt)) + geom_point() + geom_smooth(method = "lm")
## `geom_smooth()` using formula = 'y ~ x'
# Question 17 : How does the distribution of total transaction amounts vary across different card categories?
library('ggplot2')
ggplot(credit, aes(x = Card_Category, y = Total_Trans_Amt)) +
geom_boxplot(fill = c("blue","gold","purple","gray"), color = "black") +
labs(title = "Distribution of Total Transaction Amounts Across Card Categories",
x = "Card Category",
y = "Total Transaction Amount")
# Question 18 : What is the correlation between different customer attributes such as age, credit limit, total transaction amount, dependency count?
library('corrplot')
## Warning: package 'corrplot' was built under R version 4.3.3
## corrplot 0.92 loaded
correlation <- cor(credit[, c("Customer_Age", "Credit_Limit", "Total_Trans_Amt","Dependent_count")], method = "pearson")
corrplot(correlation, method = "number")
# Question 19 : What are the relationships and patterns among attrition flag, customer age, and gender in the credit card dataset?
library('GGally')
## Warning: package 'GGally' was built under R version 4.3.3
## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2
ggpairs(credit, columns = c("Attrition_Flag","Customer_Age","Gender"), aes(color = "Income_Category"))
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
# Question 20 : How does a customer's total transaction count influence their total transaction amount?
library('dplyr')
credit_data1 <- credit %>%
select(Total_Trans_Amt, Total_Trans_Ct)
lm_models <- lm(Total_Trans_Amt ~ Total_Trans_Ct, data = credit_data1)
new_total_trans_ct <- data.frame(Total_Trans_Ct = c(100, 45, 76))
predict(lm_models, new_total_trans_ct) %>% round(2)
## 1 2 3
## 8509.40 2084.13 5705.65
# Question 21 : What is the standard deviation of customer ages in the dataset?
library('dplyr')
age_std_dev <- sd(credit$Customer_Age)
cat("Standard Deviation of customer age is : ",age_std_dev)
## Standard Deviation of customer age is : 8.016814