Real-world data is rarely simple. In this module, you’ll extend regression modeling to include multiple predictors and interaction effects. You’ll explore how adding variables improves model accuracy, how to interpret complex relationships, and how to avoid overfitting as your models become more sophisticated.

Not all outcomes are numerical. In this module, you’ll learn how to model categorical outcomes (e.g., “yes/no” or “spam/not spam”) using logistic regression. You’ll discover how to calculate probabilities, classify outcomes, and assess the performance of your models. Along the way, you’ll explore how overfitting affects classification and reflect on how to interpret and communicate model predictions responsibly.

Every model comes with uncertainty and understanding and communicating that uncertainty is what makes you a thoughtful data scientist. In this final module, you’ll explore bootstrapping and randomization methods to measure confidence in your results, conduct hypothesis tests, and communicate findings transparently. By the end, you’ll bring together your modeling and inference skills to draw clear, data-driven conclusions.