ST732 - Applied Longitudinal Data Analysis

Instructor: Dr. Wenbin Lu (email: lu@stat.ncsu.edu)
Lectures: T, TH 8:30-9:45pm, 1216 SAS Hall. Syllabus 
Office Hour:
T 4:00-5:00pm, TH, 3:00-4:00pm, 5212 SAS Hall (or by appointment)

 

TAAhn, Mihye (email: mahn@ncsu.edu)          

Office Hour: 1101 SAS Hall, W 2-4pm.

Textbooks:
Lecture notes prepared by Dr. Davidian will be used. These may be purchased at the Sir Speedy across the street from Patterson on Hillsborough. You should obtain a copy.


Useful Links:

 

Data Links:

 

Data Analysis Project:

 

 

Course Schedule

Week 1

Lecture 1: Introduction for longitudinal data 

Homework 1

Lecture 2: Matrix review, multivariate normal review

 

Week 2

Lecture 3: Models for longitudinal data (Ch. 4)

 

Lecture 4: Split plot model in SAS

 

Week 3

Lecture 5: SAS glm with the repeated statement.

Lecture 6: Multivariate statistics via glm and the repeated statement and manova options (Ch. 6), restrictions of glm (Ch. 7)

Week 4

Lecture 7: Introduction to modeling longitudinal data in proc mixed, covariance structure selection

Lecture 8: "Noint" parameterization and "Difference" parameterization for dental data. Analysis for the dialyzer data, where pressure is measured at different values

 Homework 2, solution

Week 5

Lecture 9: Analysis of the hip replacement data, with missing response values, proc mixed parameters--contrast and estimate statements

Lecture 10: Units within blocks over time and comparisons with standard ANOVA

 

Week 6

Lecture 11: Random coefficient models (Ch. 9)

 

Lecture 12: Questions about Mid-term. Random cofficient models, dental data and dializer data

 

Week 7

Lecture 13: The general linear mixed model: linear terms that are either fixed or random, Best Linear Unbiased Prediction (BLUP) of random quantities

 Homework 3, solution

Lecture 14: More on BLUPS and EBLUPS. Weight lifting example

Week 8

Lecture 15: Testing of variance components

Lecture 16: Generalized linear models, independent Y's. Poisson example.

 

Week 9

Lecture 17: GLM continued. Binary and gamma examples.

Homework 4, solution

Mid-term Exam, March 11, Thursday, 8:30-9:45am

 

Week 10 

Spring Break, no class

Spring Break, no class

 

Week 11

Lecture 18: GLM continued. Binary and gamma examples.

 

Lecture 19: Iterative reweighted Least Squared (IRLS) estimation and algorithm.    

 

Week 12 

Lecture 20: Score equations and asymptotic variance

Homework 5, solution

 

Spring Holiday, no class

 

Week 13 

Lecture 21: Begin GEE, extension of GLM's to longitudinal data (Ch. 12) 

 

 

Lecture 22: More GEE    

Week 14

Lecture 23: GEE, sandwich matrix details

 

Lecture 24: Random coefficient models in ordinal logistic regression, schizophrenia example

 

Week 15 

Lecture 25: Random coefficient models in ordinal logistic regression, schizophrenia example

 

 

Lecture 26: Choosing longitudinal models--review

 

Week 16

Lecture 27: Project discussion 

 

Lecture 28: Project discussion

 

Final Exam

May 13th, Thursday, 8:00-11:00am