PORTLAND STATE UNIVERSITY

Fariborz Maseeh Department of Mathematics & Statistics

STAT578 SURVIVAL ANALYSIS
(3 credits, effective winter 2012)

Instructor:

Professor Jong Sung Kim
M307 NH
5-3657
jong@pdx.edu

Office Hours:  11:20pm - 12:10pm on MW or by appointments

Chairman:

Professor Peter Veerman
343 NH
5-3630

CRN, time, and place: 45667, MW12:45 - 14:00, SEH107

Text: Survival Analysis Using S: Analysis of Time-to-Event Data, Chapman & Hall/CRC, 2004 by Mara Tableman and Jong Sung Kim with a contribution from Stephen Portnoy; reserved at Millar Library for 2hours check-out (QA276.4 .T32 2004)

Course description: Survival analysis is the modern name given to the collection of statistical procedures which accommodate time-to-event data, having as a principal endpoint the time when an event occurs. Such events are generally referred to as failures. Some examples are time to failure of an electrical component, time to first recurrence of a tumor (i.e., length of remission) after initial treatment, time to death, time to the learning of a skill, etc. In these examples it is possible that a failure time will not be observed either by deliberate design or due to random censoring. This occurs, for example, if a patient is still alive at the end of a clinical trial period or has moved away. The necessity of obtaining methods of analysis that accommodate censoring is the primary reason for developing specialized models and procedures for failure time data.

Topics to be covered: Kaplan-Meier Estimator of Survival, Lifetime Distributions, Fitting Parametric Methods to Data, Accelerated Failure Time Models, Cox Proportional Hazards Models, Model Selection, and Model Checking including Residual Analysis and Cut Point Analysis with Bootstrap Validation.

Audience: Students from plant science, biological science, engineering and computer sciences, medical science, epidemiology and other health sciences, statistics, applied mathematics, economics, environmental sciences, social sciences, astrophysics, psychology, and demography.

Prerequisite: STAT 4/552 or consent of instructor. R experience would be helpful, but not required.

• Statistical computing package: R, free statistics software, also available at Mathematics Computer Lab, 465 NH.

Lecture

Chapter 1

Chapter2a Chapter2a_1 Chapter2b Chapter2c Chapter2c_1
Chapter3a Chapter3b Chapter3c Chapter3d Chapter3e

Chapter4a Chapter4b
Chapter4c
Chapter5a Chapter5b Cox-PH_strat-int
Extended Cox Model

Reference:

• Survival Analysis by John P. Klein, M. L. Moeschberger, Alwyn B. Scott, Melvin L. Moeschberger  (R853.S7 K535 2003)
• Survival Analysis: A Self-Learning Text (Statistics in the Health Sciences) by David G. Kleinbaum
• Analysis of Failure and Survival Data by Peter Smith
• Statistical Models and Methods for Lifetime Data by Jerald F. Lawless (QA276 .L328 2003)

Course Grade: Homework 30%; Midterm 30%; Final Project + In-class Presentation 40%

Homework will be assigned each week. Please staple together multiple pages. Students are expected to show all their work. Late homework won’t be accepted unless prior arrangements have been made.

Key#1 Key#2 Key#3 Key#4 Key#5 Key#6

TA

Handout

Paper to read paper to read II(crossing curves)

Resources such as data sets, functions, errata sheet, and updates: Click here

Final presentation:Mon 3/19/12 12:30 - 14:20