NATIONAL YANG MING CHIAO TUNG UNIVERSITY

INSTITUTE OF STATISTICS

 

LONGITUDINAL DATA ANALYSIS

SPRING 2024

 

 

 


Instructor:

Guan-Hua Huang, Ph.D.

 

Office: 423 Joint Education Hall

 

Phone: 03-513-1334

 

Email: ghuang@nycu.edu.tw

Class meetings:

Monday 13:20-16:20 at 406 Joint Education Hall

Office hours:

By appointment

Class website:

https://ghuang.stat.nycu.edu.tw/course/lda24/

Credit:

Three (3) credits

 

COURSE SUMMARY

 

Longitudinal data consist of multiple measures over time on an individual. This type of data occurs extensively in both observational and experimental biomedical studies, as well as in studies in sociology and applied economics. This course will provide an introduction to the principals and methods for the analysis of longitudinal data. While some theoretical statistical detail is given (at the level of appropriate for a Master’s student in Statistics), the primary focus will be on data analysis and interpretation.

 

The objects of his course are

 

Ÿ   To identify features of longitudinal data and explain the roles of longitudinal data in studying real data phenomenon.

Ÿ   To use a generalized linear model to make inferences about the relationship between responses and explanatory variables while accounting for the correlation among repeated responses for an individual.

Ÿ   To use marginal, random effects, or transition models for longitudinal data when the repeated observations are binary, count, or Gaussian/non-Gaussian continuous.

Ÿ   To familiarize the usage of statistical software implementing these longitudinal data analytic methodologies.

Ÿ   To provide references for your future research.

 

HANDOUTS AND TEXTBOOKS

 

Handouts corresponding to each lecture will be available on the class website before each class. Reading assignments are from the following book:

 

Ÿ   Diggle PJ, Heagerty P, Liang KY and Zeger SL (2002). Analysis of Longitudinal Data, 2nd edition. Oxford University Press.

 

PREREQUISITES

 

Students are expected to have background on undergraduate probability, and mathematical statistics. Some knowledge on (generalized) linear regression will be helpful.

 

METHOD OF STUDENT EVALUATION

 

The course grade will be based on 4 homework assignments (50%), 1 midterm exam (20%), and 1 final exam (30%).

 

COURSE OUTLINE

 

Readings refer to:

Diggle PJ, Heagerty P, Liang KY and Zeger SL (2002). Analysis of Longitudinal Data, 2nd edition. (ALD).

 

Module

Topic

Reading

1

Introduction and examples of longitudinal data

Ÿ  Introduction and examples

Ÿ  Notation for longitudinal data

Ÿ  Models for longitudinal data

ALD

Chapter 1

2

Exploring longitudinal data

Ÿ  Exploring longitudinal data

Ÿ  Exploring correlation structure of longitudinal data

ALD

Chapter 3

3

Linear modes for longitudinal data

Ÿ  Introduction, overview and simple example

Ÿ  Correlation models

Ÿ  Inferences

Ÿ  Evaluating covariance models

Ÿ  Sensitivity to covariance/correlation model and robust variance

Ÿ  Exploiting the empirical variance estimator- generalized estimating equations (GEE)

Ÿ  Where have we been?

ALD

Chapter 4

4

Linear mixed models for longitudinal data

Ÿ  Introduction

Ÿ  Linear mixed models for longitudinal data: example

Ÿ  Details of model building: inference

Ÿ  Model evaluation for linear mixed models

Ÿ  Parameterization of random effects

Ÿ  Estimating individual trajectories

ALD

Chapter 5

5

GLM for longitudinal data

Ÿ  Marginal models

Ÿ  Random effects models

Ÿ  Transition models

ADL

Chapters 7, 8, 9, and 10