A Complete 5-Day Course on the Analysis of Longitudinal Event Data
How to study the causes of
Births and Deaths
Marriages and Divorces
Arrests and Convictions
Job Changes and Promotions
Bankruptcies and Mergers
Wars and Revolutions
Residence Changes
Consumer Purchases
Adoption of Innovations
Hospitalizations
For survival data, ordinary regression analysis won't do the job
If you've ever used regression analysis on longitudinal event data, you've probably come up against two intractable problems:
1. Censoring: Nearly every sample contains some cases that do not experience an event. If the dependent variable is the time of the event, what do you do with these "censored" cases?
2. Time-dependentcovariates: Many explanatory variables (like income or blood pressure) change in value over time. How do you put such variables in a regression analysis?
Makeshift solutions to these questions can lead to severe biases. Survival methods are explicitly designed to deal with censoring and time-dependent covariates in a statistically correct way. Originally developed by biostatisticians, these methods have become popular in sociology, demography, psychology, economics, political science, and marketing.
How you will benefit from this course
Event History & Survival Analysis is a complete course covering both the theory and practice of survival methodology. In fact, it includes all the lectures and assignments in Professor Allison's one-semester course. Assuming no previous knowledge of survival analysis, this course will turn you into a knowledgeable and skilled user of these indispensable techniques. Here are a few of the skills you will acquire:
How to organize survival data.
How to choose the right time axis.
When to use discrete vs. continuous time methods.
How to handle left censoring.
How to pick the right computer program.
What to do about nonproportionality.
How to compute R-squared.
When and how to correct for unobserved heterogeneity.
How frequently to measure independent variables.
What to do if there is more than one kind of event.
How to test for sensitivity to informative censoring.
How to tell if a model fits the data
Who should attend?
If you need to analyze longitudinal event data and have a basic statistical background, this course is for you. You should have a good working knowledge of the principles and practice of multiple regression, as well as elementary statistical inference. But you do not need to know matrix algebra, calculus, or likelihood theory.
Previous participants have come from a wide variety of fields: sociology, demography, psychology, economics, management, finance, history, marketing, biology, medicine, veterinary medicine and criminal justice.
Computing
At least one hour each day is devoted to carefully structured and supervised assignments on personal computers, with one computer for each participant. Additional time is available for exploring other sample data sets. Or you can bring your own data and try out new techniques as you learn them. All assignments are done with the SAS statistical package. SAS procedures used in the course include PHREG, LIFEREG, LOGISTIC, and LIFETEST.
Location, format, materials
The course meets Monday through Friday at Temple University Center City, located in the heart of Philadelphia at 16th & Market Streets, across from City Hall.
Participants will get a copy of Professor Allison's text Survival Analysis Using SAS : A PracticalGuide. In addition, participants receive a 120-page manual containing detailed lecture notes (with equations and graphics), examples of computer printout, and many other useful features. This book frees participants from the distracting task of notetaking.
Course outline
1. Fundamentals of survival snalysis 2. Weibull and Gompertz models 3. Types of censoring 4. Life table method 5. Kaplan-Meier method 6. Accelerated failure time models 7. Maximum likelihood estimation 8. Interpretation of parameters 9. Proportional hazards models 10. Partial likelihood estimation 11. Time dependent covariates 12. Competing risks 13. Discrete time analysis 14. Sensitivity analysis for censoring 15. Choice of time origin 16. Testing the proportional hazards assumption 17. Heterogeneity and time dependence 18. R-squared 19. Repeated events 20. Left censoring, left truncation
Comments by Recent Participants
Participants in the July 2009 seminar were asked to rate the course on a scale of 1 (worst) to 10 (best). The average score for 29 respondents was 8.8. They were also asked if they wished to make an attributed statement regarding the course. Here are all the comments that were received:
“Course was a terrific tour de force through the material, leaving participants with the capability to apply these techniques to their own research interests.”
Gail Rattinger, University of Maryland
“The course provides a very useful framework of how to think about survival analysis, considering various scenarios that one might encounter. Along with discussing the theoretical background, it is very applied, which makes it even more useful. I would recommend anyone who does statistical analysis of time to event data to take this course.”
Gayane Yenokyan, Johns Hopkins University
“Great event history class. Gives the students a general as well as in-depth coverage of various event history methods! You can attend and understand the majority of the material without a general knowledge of SAS, but SAS programming skill is certainly preferable.”
Congcong Zheng, San Diego State University
“This is an outstanding environment for researchers who use statistics, but are not statisticians, to develop and augment their knowledge and improve their work. Allison does a great job relating the material to everything from education to business.”
Jacob Gross, Indiana University
“Excellent teacher, extremely knowledgeable an experienced. A great course for intermediate learning to advanced learning.”
Ashutosh Tamhane, University of Alabama at Birmingham
“This course was an incredibly valuable resource. In just one week's time, I gained a firm background in survival analysis that will allow me to focus my dissertation methods and finish my proposal. Thanks.”
Christina Andrews, University of Chicago
“This is the best survival analysis seminar or course I have attended. The instructor is well experienced and, most of all, he is very knowledgeable. I have learned a lot from his course. I highly recommend others to attend this seminar. Although the course is very intensive, the materials covered are very informative.”
Ju Helen Wong, University of Manchester
“The course makes a very clear progression from intuition to application. I wil recommend it to others in my department.”
Seth Carnahan, University of Illinois
“This course provides excellent coverage of survival analysis, both from the theoretical to practical applications.”
