arpitashrivas001 / credit_balance_analysis

Executed Regression modelling, hypothesis testing and statistical analysis to predict factors affecting credit card balances in a firm. Tools & technologies: ANOVA, p-value, R square

Credit_Card_Acitivity

SUMMARY Credit card companies are leveraging data analysis to gain a deeper understanding of the factors driving the variation in customer monthly credit card balances. The primary aim is to enhance customer service and optimize business operations through data-informed decision-making.

GOALS This study involves the utilization of regression modeling through Analysis of Variance (ANOVA) to achieve the following objectives:

1. Identification of key determinants influencing credit card activity within the organization.
2. Selection of optimal factors for accurately estimating monthly customer credit card balances.
3. Uncovering the significant factors contributing to the observed balance levels.
4. Identification and analysis of the top five factors exerting an influence on credit card balances.

PROJECT FLOW

1. Initial identification of potential variables was performed by subjecting prediction models to regression analysis.
2. Assessment of the prediction model involved evaluating metrics such as the standard error of estimate and coefficient of determination (R-squared).
3. The statistical significance of the model was established using hypothesis testing, including examination of Type 1 errors, assessment of collinearity, and analysis of p-values. Tools and Concepts: The analytical toolkit for this project includes ANOVA, predictive modeling, regression analysis, hypothesis testing, considerations of Type 1 errors, collinearity assessment, p-value analysis, determination of R-squared, and evaluation of the F-value.