IR 514: Multivariate Analysis
In the Spring of 2013 I am teaching IR 514: Multivariate Analysis for International Relations
The goals of this course are to:
1. Enable you to read quantitative research intelligently
2. Teach you several basic techniques for quantitative hypothesis testing.
3. Enable you to manage and manipulate datasets for large-n research projects
4. Provide you skills for self-teaching or working with co-authors using more sophisticated techniques in the future.
5. Provide you with a working knowledge of STATA to facilitate 2-4.
The first part of the course focuses on the derivation, estimation, and application of linear regression. The second part of the course covers violations of the assumptions of the linear regression model and methods that are used when these assumptions are violated.
The course will cover both the derivation and properties of estimators, as well as practical applications through data analysis using STATA, a statistical software package. The goal of the course is to provide students with an introduction to the tools that are necessary to conduct quantitative research in political science.
The (provisional) syllabus for the course is here (Updated January 26). Many of the readings will be updated as the semester progresses (though all of the readings currently on there will be useful).
Each week, much of the traditional lecture material will be assigned as out-of-class viewing. This saves our in-class time for Q&A, discussion, simulation, and application. Please bring your laptop to class every week so that you can run exercises on your own machine.
All .do files used for in-class simulations are in the course dropbox folder.
Week 7: See .do file in dropbox. This class was all stata.
Week 10: Multiple Regression (2)
Week 12: Autocorrelation and Panel Data
Week 13: Logit & Probit
In case I haven't harped on reading codebooks enough, here is James Vreeland's takedown of the anocracies and civil conflict work.