This is the third course in the PS 231 graduate methods sequence in the political science department. In this course, students will learn about model-based statistical inference and its applications to political science research. The course will cover multiple approaches to model-based inference. First, students will learn about maximum likelihood estimation, which proceeds by assuming the data were generated by a specified probability model. Second, students will learn a collection of methods in machine learning, which employ algorithmic models to optimize fit to the data without relying on assumptions about the data mechanism. Along the way, students will learn about the strengths and limitations of these different approaches, how to interpret the outputs of different types of models, and how to assess the value of estimated models in different situations.
Discussion scheduled Fridays 1:30-3pm.
Political Science 231A, 231B or equivalent.