I analyse the UCI credit card default dataset available at: https://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients#.
The dataset is details of credit card holders of an "important bank in Taiwan" for the period April to September, 2005. The features available include some basic customer demographics (gender, education, marital status and age), available credit line, their history of payment/default for the six months mentioned (Apr--Sep '05), their bill amounts and their payment amounts for that period and a binary target variable indicating default the following month.
The data was originally studied by the authors of the paper: Yeh, I. C., & Lien, C. H. (2009). The comparisons of data mining techniques for the predictive accuracy of probability of default of credit card clients. Expert Systems with Applications, 36(2), 2473-2480.
I perform some EDA to understand the data and clean the data, engineer relevant features, build predictive models to predict default and perform some statistical analyses to obtain a greater understanding of the features and their interactions. I finish with some business case scenarios where the predictive model could be applied.