Linear Mixed Models for Repeated Measures using R (online)
Date:
24/09/2024 - 25/09/2024
Organised by:
Statistical Services Centre Ltd
Presenter:
James Gallagher and Sandro Leidi
Level:
Intermediate (some prior knowledge)
Contact:
James Gallagher
07873873617
jamesgallagher1929@gmail.com
Venue: Online
Description:
Overview of 2-day course
In a repeated measures experiment a response variable is repeatedly measured for each subject or unit over time under the same treatment. These observations are likely to be correlated over time, rendering conventional linear model methods either inappropriate for analysis or of limited use.
Linear mixed models are commonly used to analyse repeated measurements, or longitudinal data, which are normally distributed. In this course we begin with a brief overview of repeated measures before moving onto the random coefficient model formulation of a linear mixed model (also known as subject-specific models). For the remainder of the course we focus on applying marginal models, sometimes known as covariance pattern models. Marginal models are particular useful for situations where the primary interest lies in studying mean trend through fixed effects, with variation in correlated errors about the trend treated as a nuisance.
The course will emphasise the practicalities associated with choosing, fitting and interpreting linear mixed models in the context of analysing repeated measures. Examples will be drawn from medical and health related applications.
Practical work will be based on the R software; see https://www.r-project.org. Relevant models will be fitted using the CRAN packages lmerTest and mmrm.
Cost
£534 (inclusive of 20% VAT)
Delivery Mode
All training is online and will be delivered live each day between 10:00 and 16:30 (GMT+1). Delivery platform: Zoom, which may be freely accessed. Questions may be asked using Zoom's chat box. Note our online courses are delivered by a team of two presenters, so one presenter is always available to provide additional support. We also use Zoom meetings rather than webinars to encourage further interaction during an online course.
Who Should Attend?
Data analysts and statisticians working in medicine, health and related areas who wish to have a practical introduction to the analysis of repeated measures using linear mixed models. It is assumed that participants are R users and have some familiarity the practical use of linear mixed models in general. No prior knowledge of analysing repeated measures is assumed.
How You Will Benefit
The course will give you the skills to use linear mixed models to analyse normally distributed repeated measurement data. You will also appreciate the distinction between random coefficient (subject-specific) models and marginal models, and their advantages and disadvantages.
What Do We Cover?
- Overview of repeated measures
- Random coefficient models; lmerTest CRAN package
- Marginal models and covariance structures
- Fitting marginal models using the mmrm CRAN package
- Random coefficient models versus marginal models
- Model selection for marginal models
- Inferential methods; Kenward-Roger for fixed effects and likelihood ratio testing and AIC for covariance structures
- Model checking for marginal models
- Further complexities associated with the analysis of repeated measures, e.g. relationship between random coefficient and marginal formulations of the mixed model, negatively correlated repeated measures data, convergence issues.
The course does not cover GEE type models.
Software
Practical work will be done in R.
Note: For practical work, participants must download and install a number of CRAN packages in R. This must be done prior to the start of the course.
Cost:
£534
Website and registration:
https://www.ssc-training.co.uk/linear-mixed-models-for-repeated-measures-using-r.html
Region:
International
Keywords:
Quantitative Data Handling and Data Analysis, Multilevel Modelling , Hierarchical models, Mixed models, Random effects, Longitudinal Data Analysis, Random coefficients, Marginal models, Covariance patterns, mmrm, Kenward-Roger, Satterthwaite, Model checking, Scaled residuals, Transformed, Normalized, Covariance structures, Unstructured, Compound symmetry, Autoregressive
Related publications and presentations:
Quantitative Data Handling and Data Analysis
Multilevel Modelling
Hierarchical models
Mixed models
Random effects
Longitudinal Data Analysis