GOV 380R • Math Methods for Political Analysis
12:30 PM-2:00 PM
In this course we will study selected advanced topics in political methodology, including generalized regression models, models with categorical or limited dependent variable, event count models, event history models, and time-series cross-section models. Most of these models rely on the maximum likelihood method for estimation. In order to understand the maximum likelihood principle, we will first introduce matrix algebra and vector/matrix differentiation. We will use STATA for statistical analysis and MAPLE for symbolic algebra.
You are required to write a substantive paper based on a statistical procedure that employs maximum likelihood estimation. The topic of the paper is your own choice, but you should discuss your ideas with the instructor early in the semester to get his approval. Depending on substantive merits, topics based methods introduced in a lower level course may not be acceptable. You should work closely with the instructor in developing ideas, formulating models, acquiring data, and carrying out the analyses. Your grade will be based on the end result as well as your interactions with the instructor working on this paper. There may be homework assignments and/or exams at the instructors discretion.
1. W. H. Greene, Econometric Analysis (5th Ed.). Prentice Hall, 2003. 2. J. M. Box-Steffensmeier and B. S. Jones, Event History Modeling: A Guide for Social Scientists. Cambridge, 2004. 3. Others to be determined.