Applied Multivariate and Longitudinal Data Analysis (495/590 - 037 Spring 2016)

                    TuTh 10:15AM - 11:30AM   SAS HALL 1216


Professor:  Dr. Ana-Maria Staicu [last name pronounced as "styku"]

Contact Information
Office: 5220 SAS Hall
Phone: 919 515-0644
Email: astaicu [at]  ncsu [dot] edu 
Office Hours
Tuesday 12:00PM-1:30PM 

Teaching Assistant: Saebitna Oh
   Office Hours: Monday 1-2:30PM / Wed 10-11:30AM Room: 1101 SAS Hall
   Email: soh3 [at] ncsu [dot] edu

Software used in this course: R is freely available at

    Download and install R. Go to and follow the instructions at the top of the page.
    UCLA R Tutorial


Grading: Homework – 40%, Midterm – 30%, Final exam – 30%

Grades so far [Final grade is out of 85]

Project description

Groups for the Group Project  [posted on March 17 @5:30PM]

Sources for data:

Center for the study of aging and human development
American Psychological Association

Feb 2: Class CANCELLED. Also Office Hours are CANCELLED

Lecture Notes

Ch 1. Introduction to Basic Concepts
    What are multivariate/longitudinal data ?
    Review: matrix algebra   Additional Notes [reference: ]
    Review: common distributions

    HW1 (typos corrected)  Solutions

    Review: Random Variables. Sampling distribution.

Part 1: Applied Multivariate Analysis

Ch 2. Inference about mean vectors
    Inference for a single population/multiple populations . R Illustration:  iris data example
R Illustration:  Examples2  T6-1  T6-2
    HW2   Datasets: Table 5.12  Table 5.13    Solutions

    R illustration : Example3 T6-4.dat   T6-14.dat
    HW3   Datasets: Table T6-12.dat  Table T6-13.dat   Solutions
    Datasets (.dat)

Ch 3. Principal components analysis
    Principal Component Analysis  [Last updated 6:40PM 02/08] R illustration: iris example2
    ReturnRates example    Data
    HW 4  Dataset: Table 8.4   

Ch 4. Factor analysis
    Factor Analysis.
[Last updated 6:50AM 02/15]  R illustration
    HW 5  Dataset:  Table 9.12   Solutions

Ch 5. Classification and clustering
Discrimination and classification  [Last updated 9:10AM 02/25] R illustration: iris example3
    R illustration: Toy example and Examples     Mower data

Midterm [Thursday, March 3 @ 10:15AM]

Part 2: Applied Longitudinal Data Analysis

Ch 6. General linear model
    Intro to modeling longitudinal data  [
Last updated 5:00PM 03/21

           R example1  dental data
           R example2 Vlagtwedde-Vlaardingen data (description @

           Exploratory Analysis || Fitting Marginal Model   [geeglm] ||
Fitting Marginal Model 2  [gls]

           ggplot cheatsheet (courtesy of R studio)
    HW 6  [due on Monday, March 28 @1PM, 1101 SAS Hall ] Dataset:  insulin.txt   Solutions     

    HW 7 One-page proposal [one submission per group]   Template

Ch 7. Linear mixed model (LMM)
   Linear mixed model
[Last updated 9:30AM 04/07LME4 vignettes

Fitting Mixed Effects Model [lmer] Six Cities Air Pollution data

    HW 8  [typo corrected] [HW8 due on Thursday, April 7]  Dataset: lead data   Solutions
Ch 8. Generalized linear model (GLM)
   Review of GLM for scalar response
[Last updated 8:00AM 04/11]
   HW 9[HW9 due on Wednesday April 20 @10AM in 1101     SASHall ]  Exercise data 

Generalized linear model for repeated responses GEE and GLMM  [Last updated 7:30PM 04/18]
   Data sets: Epileptic seizures  ||  Wheezing

   GEE analysis of epileptic seizures
   GEE analysis of  wheezing
Final Exam:
The final exam comprises: a written exam portion [
Thursday, April 21] + group final project report [Thursday April 28].
A proposal for the group final project is due on Thursday March 31.

Date Posted

  • 01/04/16 First Day of Class: Thursday Jan 7. Last day of class Thursday April 21.