Credit Default Prediction with Python

This course provides a hands-on journey into credit risk prediction using Python with a focus on logistic regression, decision trees, and ensemble methods. Learners will begin by outlining project workflows, importing data, and applying data preprocessing techniques such as handling missing values, encoding categorical features, and scaling numerical variables. Through exploratory data analysis (EDA), they will interpret data patterns and relationships to build stronger foundations for modeling. Moving into advanced modeling, learners will evaluate models using confusion matrices and ROC curves, ensuring accuracy and reliability in predicting defaults. They will optimize logistic regression models through hyperparameter tuning methods like Grid Search and Randomized Search. Expanding further, the course introduces decision tree theory and practical coding steps, enhanced with visualization using Graphviz for interpretability. Finally, learners will construct Random Forest models to reduce overfitting and improve predictive performance, applying ensemble learning techniques to real-world credit datasets. By the end of this course, learners will be able to apply, analyze, evaluate, and construct predictive models that enhance decision-making in financial risk management, using industry-standard tools and Python libraries

• Preprocess financial datasets using encoding, scaling, and exploratory data analysis
• Build and tune logistic regression, decision trees, and random forest models
• Evaluate credit risk models using confusion matrices, ROC curves, and ensemble methods

• Basic computer and internet skills
• Ability to read course instructions in English and complete short activities