Biostatistics 236 Modeling Longitudinal Data

Spring 2023 - Robert Weiss

News & Announcements:

• Request: Attend class in person (unless you're ill of course). It's a little bit odd to be lecturing to between 3% and 10% of enrolled students. 
• Suggestion: Wear a mask. Then we don't share infectious diseases (flu, cold, RSV, Coronavirus, and all the other named and unnamed infectious diseases that we share by breathing). 
• SAS Studio Introduction (word) Please follow instructions on first page and register for SAS Studio. Ideally immediately, but definitely before lab 8 Wednesday May 24th.
• Zoom links Office Hours and Class for 2023 (in pdf).
• We will be moving to R for computer labs. Conversion is a major effort, please bear with us during this move. If you find errors or know of easier/better ways to do some code, please let me know. 
• Data Analysis Project Comments and General Advice.
• Variance Covariance Choices.
• Let me know if any errors.
• 2023 Syllabus has been posted.

Table of Contents

• Class Schedule
• Syllabus and Textbook
• Lecture Notes
• Homework 
• Computer Labs in R
• Computer Labs in SAS
• Data Sets
• Discussion Questions
• Topics from lectures and labs 
• Zoom recordings of lectures/labs are available in the Media Gallery on Bruin Learn. 
• Zoom links Office Hours and Class for 2023 (in pdf).

2023 Class Schedule, Email Addresses, and Office Hours:

For Zoom links see 2023 Course Zoom links

Syllabus and Textbook

2023 Syllabus has been posted. Includes 2023 HW due dates and 2023 grading system.

The 2023 textbook is Modeling Longitudinal Data. The book web page has data and links to example code.

The textbook is available for download through the UCLA library Springer bookshelf. If you are off campus, you will need to VPN to campus to be able to download the book. When I type "Modeling Longitudinal Data" (without the quotes) into the search box at UCLA Library webpage, the book is the first item on the list.

Lecture Notes

• Zipped folder of all 9 course note packets but don't have the minor edits. Password Protected.
• Lecture notes 1 intro.
• Lecture notes 2 graphics.
• Lecture notes 3 simple critique MVN.
• Lecture notes 4 predictors.
• Lecture notes 5 covariance.
• Lecture notes 6 random effects.
• Lecture notes 7 examples missing computing.
• Lecture notes 8 computation examples discrete.
• Lecture notes 9 bivariate.

Variance Covariance Choices.

Homework Assignments

Homeworks (PDFs)

Please turn in PDFs of your solutions on Bruin Learn. Due dates are on the syllabus.

All data Sets in both Text and SAS Formats

• Homework Number 1 
• Papers for HW #1 problem 1
  • Counsell et al (2007)
  • HLP (2007)
  • Ponsford et al (2014)
  • Rabiolo et al (2020)
  • Li et al (2021)
• Homework Number 2
• Homework Number 3
• Homework Number 4
• Homework Number 5
• DAP1 Macular Data Analysis
  • Macular Data SAS file
  • Macular Data Excel file
  • Macular Data .csv file
  • Macular Data .txt file
• Final Data Analysis Project (FDAP).

Homeworks (Original LaTeX)

• Homework Number 1
• Homework Number 2
• Homework Number 3
• Homework Number 4
• Homework Number 5
• DAP1 Macular Data Analysis
• Final Data Analysis Project (FDAP)

Computer Labs in R for Learning Longitudinal Data Analysis

• Lab 1 Data Management and Introduction
  • RLab 1 Intro.docx
  • Lab1.R
  • Pediatric Pain data (text file)
  • Cognitive Data Set for TODO portion of lab.
• Lab 2 Graphics
  • RLab 2 graphics.docx
  • Lab2_graphics.R
  • bigmice data
  • smallmice data
• Lab 3 Simple Analyses + 1 Longitudinal Random Effect Model
  • Lab3_simple_analyses.docx
  • Rlab3 _simple_analyses.R
  • Pediatric Pain data (text file)
• Lab 4 Longitudinal Models for the Pediatric Pain Analysis. 
  • R Output
  • R Commands
  • Pediatric Pain data (text file)
• Lab 5 Covariance Models for the Small Mice Data 
  • Handout
  • R Commands
  • Selected R Output
• Lab 6 Hierarchical Models BSI data and Weight Loss Data
  • Handout
  • R Commands
  • Selected R Output
• Lab 7 Longitudinal Models for Discrete Data
  • Handout
  • Selected Output
  • R Commands
• Lab 8 Using SAS to fit Covariance Models
  • SAS Lab 5 (word)
  • SAS Lab 5 selected output (word)
  • SAS Lab 5 Questions and Answers (word)
• Lab 9 Residual Analysis for Random Effects Models
  • Lab 9 (word)
  • Pediatric Pain SAS File
  • Weight Loss SAS File
  • SAS output, word document (warning: 56 pages long!) (word)

Computer Labs for Learning Longitudinal Data Analysis Using SAS

• Lab 1 Initial Exploratory Data Analysis
  • SAS Studio Introduction (word)
  • Lab 1 (word)
  • Pediatric Pain data (text file)
  • Cognitive Data Set for TODO portion of lab.

• Lab 2 Graphics
  • Lab 2 (word)
  • Pediatric Pain SAS data set for use in Lab 2. (word)
  • Mice Big (Balanced with missing format)
  • Mice Small
• Four needed Macros (written by Michael Friendly)
  • boxanno.sas
  • gdispla.sas
  • gensym.sas
  • scatmat.sas
• Lab 3 Simple Analyses
  • Lab 3
  • Lab 3 selected output
  • Pediatric Pain Sas data set.
  • Example code for Dental Data
• Lab 4 Fixed Effects: Class variables and estimate, contrast and lsmeans statements
  • Lab 4(word)
  • Lab 4 comments (word)
  • Pediatric Pain Text File
• Lab 5 Covariance Models
  • Lab 5 (word)
  • Lab 5 selected output (word)
  • Lab 5 Questions and Answers (word)
• Lab 6 Hierarchical/Nested Data
  • Lab 6 (word)
  • Selected output and comments (word)
  • BSI Total
  • Weight Loss 2 (day of observations, visit type)
• Lab 7 Discrete Data
  • Lab 7 (word)
  • NLMixed and Glimmix Malaria output (word)
  • SAS Glimmix Procedure (links at bottom for the procedure and for the documentation).
  • Morbidity (Kenya)
• Lab 8 Bivariate Longitudinal Data
  • Lab 8 (word)
  • Pediatric Pain Text File
  • Selected output and comments 1 of 2 (word)
  • Selected output and comments 2 of 2 (word)
  • Lecture notes on bivariate longitudinal data
• Lab 9 Residual Analysis for Random Effects Models
  • Lab 9 (word)
  • Pediatric Pain SAS File
  • Weight Loss SAS File
  • SAS output, word document (warning: 56 pages long!) (word)

Data Sets

• Data Sets in both SAS and Text Format

Discussion Questions

• Discussion question 2.
• Discussion question 1.

Lecture Recordings

Lecture recordings posted to our BruinLearn Video page.

• Syllabus (non-lecture, pdf).
• Lecture 01 Monday 2023_04_03. Syllabus. Lecture notes 1 intro pages 1-12. Lab 1 data sets. 
• Lecture 02 Wednesday 2023_04_05. Lecture notes 1 pages 12-17. 
• Lecture 03 Monday 2023_04_10. Lecture notes 1 pages 17- 28. Lecture notes 2 Graphics pages 1-38. 
• Lecture 04 Wednesday 2023_04_12. Lecture notes 2 Graphics pages 38-47. 
• Lecture 05 Monday 2023_04_17. Lecture notes 2 pages 48-49. Lecture notes 3 simple methods pages 1-17. Lecture notes 3, multivariate normal model pages 18-19. 
• Lecture 06 Wednesday 2023_04_19. Lecture notes 3, multivariate normal model pages 20-31. 
• Lecture 07 Monday 2023_04_24. Lecture notes 3 pages 31-48. Lecture notes 4 specifying predictors pages 1-10. Note: First hour not recorded, apologies. 
• Lecture 08 Wednesday 2023_04_26. Lecture notes 4 pages 8-10. Handout of annotated output for lab 4, pages 1-7. 
• Lecture 09 Monday 2023_05_01. Lecture notes 4 pages 11-26 (top of page 26). RLab 5 selected output. 
• Lecture 10 Wednesday 2023_05_03. Lab 5 covariance modeling.
• Lecture 11 Monday 2023_05_08. FDAP mention. Lecture notes 4 pages 26-40. Lab 6. 
• Lecture 12 Wednesday 2023_05_10. Plots from lab 6. Lecture notes 4 pages 41-53. 
• Lecture 13 Monday 2023_05_15. Output from lab 7. Lecture notes 8 pages 14-35 on discrete outcomes. Lecture notes 4 pages 53-61.  
• Lecture 14 Wednesday 2023_05_17. Finished lecture notes 4 covariates. Started lecture notes 5 covariance modeling pages 1 to 10. Sound was off for a likely significant portion of the lecture, very sorry. 
• Lecture 15 Monday 2023_05_22. Lecture notes 5 covariance modeling, pages 11 to 30. Lab 8, which is the covariance model fitting lab 5 from the SAS modelings. 
• Lecture 16 Wednesday 2023_05_24. Extended lab introducing SAS proc mixed. 
• Lecture 17 Monday 2023_05_29. Memorial Day Holiday
• Lecture 18 Wednesday 2023_05_31.Lecture notes 5 covariance modeling, pages 35-42, Lecture notes 6, random effects modeling pages 1-3, also brief discussion of ADAP, and discussion of residual analysis in lab 9. 
• Lecture 19 Monday 2023_06_05. Finished lectures notes 6, pages 3-33. 
• Lecture 20 Wednesday 2023_06_07. Lecture notes 7, Random effects models, Missing Data, pages 1 to 7. 

Labs.

• Lab 01 Wednesday 2023_04_05. Introduction, Data Management
• Lab 02 Wednesday 2023_04_12. Graphics
• Lab 03 Wednesday 2023_04_19. Simple Analyses
• Lab 04 Wednesday 2023_04_26. Fitting Longitudinal Models
• Lab 05 Wednesday 2023_05_03. Covariance Models
• Lab 06 Wednesday 2023_05_10. Clustered Data
• Lab 07 Wednesday 2023_05_17. Discrete Data
• Lab 08 Wednesday 2023_05_24. Using SAS to fit covariance models
• Lab 09 Wednesday 2023_05_31. Residuals (in SAS)
• Lab 10 Wednesday 2023_05_07. No lab, informal office hour.

SAS Documentation

• Code examples for all of Modeling Longitudinal Data!
• On line Proc Mixed documentation from SAS Institute
• SAS Glimmix Procedure and Documentation