Course Description

This course offers a rigorous and hands-on introduction to econometrics, a sub-discipline of economics that deals with quantitative analysis of economic data. The course will start with a refresher in probability and statistics and move on to ordinary least squares regression with one and several regressors, the treatment of non-linear functional forms, a critical assessment of internal and external validity in empirical work and finally ridge and lasso regression a method used by machine learning to avoid overfitting linear models.

The course follows the structure of Stock and Watson’s Introduction to Econometrics. We will use applications to give a foundation to theoretical underpinnings.  Almost each session includes a real dataset drawn from the textbook and a software lab. We use EViews as our predominant software teaching tool with regression diagnostics, and Python on Google Colab for flexibility, reproducibility, and integration with agentic AI coding assistants. Students should be comfortable moving fluently between both environments by the end of the term.

The course is built around a workflow applied to every problem: pose a question, locate appropriate data, specify a model, estimate, run diagnostics, interpret carefully, and communicate findings. While the algebra is important this mental workflow approach will be an important takeaway.