Soc
884.09
Event History Analysis
Lecture: Monday and Wednesday, 1:30-3:18, SIL
Professor:
Zhenchao Qian. Office is 346 Bricker Hall. My office hours are 12:30 to 1:30 on Monday and Wednesday. My telephone number is 688-8612, and my email is qian.26@osu.edu. My webpage is http://www.sociology.ohio-state.edu/zcq.
Overview:
An application-oriented course on statistical methods for the analysis of longitudinal data on the occurrence of events, such as births and deaths, marriages and divorces, arrests, convictions, and recidivism, career changes and promotions, wars and revolutions, residential moves, foundings and bankruptcies, etc. We will learn regression models in which the risk of event occurrence is a function of a set of time-constant and/or time-varying explanatory variables. Topics will include censoring, life tables, parametric models, Cox models, discrete-time models, unobserved heterogeneity, and repeated events.
Required Readings:
Box-Steffensmeier, Janet M. and
Bradford S. Jones. 2004. Event History Modeling: A Guide for Social
Scientists. Cambridge (B&J)
Paul D. Allison. 1995. Survival Analysis Using the SAS System: A Practical Guide. Cary, NC: SAS Institute. (A)
Recommended Readings:
Ulrich Kohler and Frauke Kreuter. 2005. Data Analysis Using Stata. Stata Press.
A comprehensive introduction to Stata with an emphais on data management, graphs, and regression.
Cleves, Mario A., William W.
Gould, and Roberto G. Gutierrez.
2004. An Introduction to Survival Analysis Using Stata. Revised Edition. Stata Press. (CGG)
Exam and Homework:
There will be one exam (35% of the grade) and five computing assignments (20% of the grade).
Project:
You are required to turn in a paper on or
before December 7. A typical paper will
report on analysis of some event history data set chosen in consultation with
the instructor. Although I encourage
you to have an independent project, it is acceptable that two or three students
analyze one data set together. However,
you are responsible for your own project.
The writing of the project must be independent (35% of the grade).
Presentation:
We will read several published articles related to event history analysis. You (two or three as a team) will be responsible for presenting one of the articles. I will prepare a list of articles. You can either choose one from the list or elsewhere (with my approval if elsewhere). I will then schedule your presentation. Each team must distribute their article to the class at least one week prior to their scheduled presentation. All students MUST read each article and be ready for in-class discussions. (10% of the grade)
Schedules:
I shall try to adhere to the following schedule. All readings should be done prior to the date of the class for which they are assigned. The exception is the reading for September 21. Please do the reading for this date and the September 26 class by September 26.
|
Date |
Topic |
Reading |
|
9/21 |
Introduction, Math
and Probability Functions Reviews |
A: Chapter 1 and Chapter 2: 14-22; B&J: Chapter 1 |
|
9/26 |
Event History Data
Structure, Censoring |
A: Chapter 2: pp. 9-14, pp. 22-27; B&J: Chapter 2 |
|
9/28 |
Descriptive Methods: Kaplan-Meier and Life Table |
A: Chapter 3 |
|
10/3 |
||
|
10/5 |
||
|
10/10 |
Stata |
CGG |
|
10/12 |
Parametric
Regression Models |
A: Chapter 4; B&J: Chapter 3 |
|
10/17 |
||
|
10/19 |
||
|
10/24 |
Cox Regression
Models |
A: Chapters 5 & 6; B&J: Chapters 4, 6, 7, 8 |
|
10/26 |
No Class |
|
|
10/31 |
Cox Regression
Models |
|
|
11/2 |
||
|
11/7 |
||
|
11/9 |
Discrete Time
Models (I) |
A: Chapter 7 B&J: Chapter 5 |
|
11/14 |
Discrete Time
Models (II) |
|
|
11/16 |
Further Issues in
Non-repeated Events |
|
|
11/21 |
Heterogeneity and Time Dependence |
A: Chapter 8; B&J: Chapter 9 |
|
11/23 |
Project
Consultation |
|
|
11/28 |
Panel Data and Repeated Events |
B&J: Chapter 10. |
|
11/30 |
Event History Analysis in Social Science Take Home Exam |
B&J: Chapter 11 |