Mixed Models in Julia

This repository contains Jupyter notebooks of chapters in a book called Mixed-effects models in Julia. Well, at least that's the plan. I'm better at starting books than I am at finishing them so maybe this will just end up being a series of notebooks.

At present the order of the chapters is

1. SimpleLMM
2. MultipleTerms

Other notebooks in the parent directory and in the CaseStudies directory may become appendices or may be incorporated in later chapters.

The plan is that the notebooks can be run on JuliaBox. Log in to the site using one of the available login methods and create a copy of this repository under the Sync tab. Some of the Julia packages used in a notebook may need to be installed. Check with

Pkg.update()
Pkg.status("MixedModels")

If Pkg.status does not return a version number, add the package with

Pkg.add("MixedModels")

Because of dependencies, MixedModels will also cause DataFrames, GLM, and Distributions to be installed.

At present R is available on JuliaBox for the Julia RCall package but the lme4 package for R is not. It can be installed with the Julia command sequence

Pkg.add("RCall")
using RCall
R"""
install.packages("lme4", repos="https://cloud.r-project.org")
"""

but this takes a bit of time and the lme4 package will not persist between JuliaBox sessions. In other words, this must be done for each fresh JuliaBox session.

An alternative is to clone this repository, install Julia (see the downloads page, and R, start julia and execute

Pkg.update()
Pkg.add("IJulia")  # this takes a while the first time it is run
Pkg.add("MixedModels")
Pkg.add("RCall")
using RCall
R"""
install.packages("lme4", repos="https://cloud.r-project.org")
"""
using IJulia
notebook()

This will open a browser window. Navigate to the clone of this repository and click on the first notebook. See the Help tab for instructions on how to use a notebook.