Essex Summer School in
Social Science Data Analysis

Matt Golder

Matt Golder is an Associate Professor in the Department of Political Science at Pennsylvania State University. He serves on the Macro Variables Committee for the Comparative Study of Electoral Systems (CSES) project and has recently been an associate scholar for the Empirical Implications of Theoretical Models (EITM) program at UCLA (2007). His recent publications have appeared in the American Journal of Political Science, British Journal of Political Science, Political Analysis, Comparative Political Studies, Electoral Studies, and European Union Politics. He has also co-authored a comparative politics textbook, Principles of Comparative Politics, by CQ Press (2008).

2L Introduction to Duration Models

Matt Golder, Florida State University
23 July - 3 August (two week course / 35 hrs)

THIS COURSE IS NOW FULLY BOOKED AND WE ARE OPERATING A WAITING LIST

Detailed Course Outline [PDF]

Course Content

This course introduces students to duration models. These models are sometimes also referred to as survival or event history models. The goal of these models is to analyse the duration of time until some event happens. The event in question might be something like the collapse of a government, the onset of war, the emergence of democracy, the adoption of a policy, the loss of a job etc. The course will be divided into three main sections:

1. Continuous Time Duration Models: We will examine parametric duration models (exponential, weibull, log-logistic, generalized gamma etc.) and semi-parametric duration models (Cox model). In addition to seeing how these models are estimated and interpreted, we will also look at various residual-based diagnostic tests.
2. Discrete Time Duration Models: We will look at the connection between discrete time duration models and binary time-series-cross-section models. We will examine various ways to deal with time dependence, ongoing events, multiple events, and time varying covariates. Finally, we will take a look at Markov transition models.
3. More Advanced Duration Models: We will examine more advanced duration models dealing with competing risks, split populations, heterogeneity, frailty, and repeated events.

Before starting these three sections, we will briefly discuss simulation methods to help calculate quantities of interest.

Course Objectives

The central objective of this course is to learn how to identify, and correctly apply, the statistical techniques appropriate to answering social science questions relating to events and duration. We will devote considerable time to computing, and conveying, quantities of interest based on model parameters. Students will be required to calculate all quantities of interest manually (not using pen and paper!) before employing the automatic commands common to statistical packages. By the end of the course, students should be quite adept at programming STATA to estimate and interpret a wide variety of different duration models.

Course Prerequisites

This is not an introductory course. Students should already have some experience with the theory behind Maximum Likelihood Estimation. Some knowledge of basic calculus (differentiation and integration), exponents and logarithms, and STATA code will be helpful.

Background Reading

King, Gary, James Alt, Michael Laver & Nancy Burns. 1990. “A Unified Model of Cabinet Dissolution in Parliamentary Democracies.” American Journal of Political Science 34: 847-871.

Diermeier, Daniel and Randy T Stevenson. 1999. “Cabinet Survival and Competing Risks.” American Journal of Political Science 43: 1051-1098.

Box-Steffensmeier, Janet M. & Christopher J. W. Zorn. 2001. “Duration Models and Proportional Hazards in Political Science.” American Journal of Political Science 45: 972-988.

Required Reading

Box-Steffensmeier, Janet & Bradford S. Jones. 2004. Event History Modeling: A Guide for Social Scientists. New York: Cambridge University Press.

[top of page]