Nardos Solomon

Edit Content

Nardos Solomon

info@nardossolomon.com
Let’s connect and explore! Whether you have a specific project in mind or simply want to chat about the fascinating intersection of data and life, I’m here. Let’s make data-driven magic together!
Linkedin Medium Github
  • Project

Credit Card Company Customer Data Analysis

The project demonstrates the use of R programming for exploratory data analysis and regression modeling in the context of credit card customer data. By leveraging R’s capabilities, the analysis contributes to customer profiling, risk assessment, strategic decision-making, and operational efficiency within the credit card company. The incorporation of statistical techniques and visualization tools in R enhances the overall understanding of customer behavior and financial dynamics.
This project aims to analyze customer data from a credit card company, focusing on variables such as age, education years, union membership, employment length, household income, debt-to-income ratio, other debt, and credit card debt. The primary goal is to understand the relationships between these variables and build a regression model to predict credit card debt.

Skills Used:
– Data Cleaning and Exploration: Utilized tidyverse functions to handle duplicate and missing values, and explored the data using summary statistics, histograms, and boxplots.

– Data Visualization: Created visualizations using ggplot to represent the distribution of numeric variables and the distribution of categorical variables like gender and union membership.

– Outlier Detection: Employed boxplots to identify and remove outliers in the CreditDebt variable.

– Correlation Analysis: Calculated and visualized correlations between numeric variables using ggcorrplot.

– Scatterplot Analysis: Utilized scatterplots to visually assess relationships between CreditDebt and other numeric variables like DebtToIncomeRatio and HHIncome.

– Linear Regression Modeling: Built linear regression models to investigate relationships between CreditDebt and independent variables, starting with HHIncome and later including OtherDebt and DebtToIncomeRatio.

– Log Transformation: Explored the impact of log transformation on the dependent variable (CreditDebt) to improve model fit.

– Model Evaluation: Assessed model assumptions using diagnostic plots (Residuals vs Fitted, Normal Q-Q, Scale-Location, Residuals vs Leverage).

Hire Me
Other Projects