Description: This repository offers a comprehensive analysis of customer credit data, aiming to provide valuable insights for financial decision-making. The project employs Python and relevant data science libraries to conduct exploratory data analysis (EDA) on the dataset. Below are the key steps performed:
Import Relevant Libraries: The analysis begins by importing essential Python libraries including Pandas for data manipulation, Matplotlib for data visualization.
Bivariate Plot for Correlation: A bivariate plot is created to investigate the potential correlation between credit card limits and the average purchase amounts made on the card. This plot aids in understanding the relationship between these two variables.
Distribution Visualization and Outlier Detection: Visualizations are generated to display the distributions of credit card limits and average purchase amounts. Additionally, outlier detection techniques are employed to identify any unusual data points that may impact the analysis.
Income Group Analysis: Using a bar chart, the distribution of customers across different income groups is visually represented. This provides insights into the income distribution of the customer base.
Total Transaction Amount Frequency Distribution: A histogram is plotted to illustrate the frequency distribution of the total transaction amounts. This helps in understanding the spread and concentration of transaction values.
Customer Retention Visualization: A pie chart is created to graphically represent the percentage of customers who have been retained versus those who have attrited. To emphasize the attrited customers, they are visually sliced apart from the main pie.