I will try my best to make this page as accurate as possible but any changes in assignments, due dates, or anything else reported in class will take precedence, whether they are reflected in these pages or not. Your responsibilities are as they are reported in class.

Fall, 2016      STA 635 - 001  Survival Analysis  [survivalability and life testing]

Do you know which paper in all statistical science that are most cited?   (Answer: the Kaplan-Meier 1958 paper)

Do you know which paper in all statistical science that are 2nd most cited?   (Answer: the Cox 1972 paper)

We shall cover the topic of both papers in this course, and more.

What is Survival Analysis? [what makes Survival Analysis special?]

   Someone said: "Ordinary statistics concerns with CDF F(t), while survival analysis concerns the survival function S(t)" [But that is a minor difference.]    In my opinion, Survival Analysis has these unique features
(1) In survival analysis, observations are subject to censoring.
(2) Nonparametric procedures are more popular, since there is no clear choice of a parametric family of distributions (like normal distributions).
(3) Survival analysis focuses on the parameter of  hazard, or hazard functions.
(4) Survival analysis is Kaplan-Meier estimator + logrank test + Cox regression model.

Instructor: Dr. Mai Zhou

Office: MDS 343, Phone: 257-6912, E-mail: mai@ms.uky.edu, Web page (this page): http://www.ms.uky.edu/~mai/sta635.html
Office Hours: TBA or by appointment.

Class:  Monday, Wednesday and Friday 1:00pm - 1:50pm at MDS 337. 

Textbook: Survival Analysis Using SAS   Second Ed. by Allison. 
                   plus my Lecture Notes. (listed below and will list more as we  procede)

Homework   Homework1   Homework2  Homework3  Homework4    Midterm Exam: Oct. 19  Homework5   Homework 6       Final Exam

My Lecture Notes for this class: 

A review of exponential distributions
Independent Censorship Model
Likelihood for Censored Data and some further explaination Note
A review of likelihood based inference methods (Chapter 10 of This Lecture Note ) and A short summary
Cox proportional Hazards Regression Model PDF file
Survival Analysis with software R PDF file
Log rank test is a video game

SAS manual for lifereg
SAS manual for Lifetest
Data sets from the book "Survival Analysis" by Klein and Moeschberger
SAS program and Data set from our textbook, 1st Ed. For second Ed. See here.
Testing related to Medians

Examples:  
 Examples of Wilks confidence interval
 More Examples    
 Wilks example
More Examples
Four SAS examples
An R package: rankreg.zip
MillerHalpern82
Time dependent covariate in Cox model by Fisher and Lin 1999
SAS for Nelson-Aalen Est.
Example illustrate invariance of Wilks Confidence interval and as a comparison see this too.

What this course is NOT: this is not a course on SAS programming (even though we use a textbook that
says ...using SAS). This is not a course about R programming either. You should know both before taking
this course. This is not a theoretical course either, I would like to think of it as a methodology course. There will be
plenty of heuristics, hints will be given to help understand statistical methods, and how to approach to prove it.
Some results are just given and not proved (or only proof in a simpler special case), or hand-waving where rigour is needed.


Some Additional Reading Materials (some at a higher level):
The Nelson-Aalen estimator and Kaplan-Meier estimator are NPMLE Note1
Learn Counting Process in 25 Minutes!
Kolmogorov-Smirnov Test
My notes  (on counting processes).
Notes on AFT model and empirical likelihood will be distributed in class.
Empirical likelihood with censored data
LASSO for Cox regression

Takehome  Final



Weekly Topics:
Exponential distribution and extensions (Weibull, extreme value, piecewise exponential).
Hazard functions. A review of  likelihood based inference methods. Likelihood in terms of hazard.
The Nelson-Aalen estimator.
The Kaplan-Meier estimator. Properties of Nelson-Aalen and Kaplan-Meier estimator.
Inference. Greenwood formula.
The Log rank test. Generalizations.
The Cox proportional hazards regression models.
The AFT models.
 

Evaluations:
Homework  50%
Midterm Exam 25%
Final Exam 25%   

Make-up Policy for Missed Exams: Make-up quizzes and exams will be given only for university excused absences. Requests must be made at least one week prior to the exam, when possible, and must be approved. If you are unable to attend and exam due to unforeseen circumstances you must contact me or the department office (257-6115) as soon as possible (within 2 days). Absences due to illness must be documented by a clinic, doctor or hospital visit and a note of explanation. Late homework and computer assignments will be accepted only for university excused absences.