Mixed Effects Regression models for LInear and Non-linear outcomes

The merlin packages implement the methodology described in the paper:

Crowther MJ. Extended multivariate generalised linear and non-linear mixed effects models. arXiv:1710.02223.

You can get the preprint here, which I recommend you read before trying out the accompanying software (note, merlin used to be called megenreg, and the syntax is out of date in the pre-print!). The software is being constantly developed and improved. This page details the development of the package, both in terms of version histories and planned features to be added. Currently merlin is only available in Stata, but we are working on the R version.

Please get in contact if you find it useful in your methodological or applied work. And of course, please cite the above paper.

Links to package specific tutorials illustrating the many uses of the framework can be found below.

$~$

### merlin in Stata

An introductory preprint on Stata’s merlin (submitted to the Stata Journal) can be downloaded here. The current stable version of merlin can be installed from the SSC archive by typing:

ssc install merlin


Any updates can be picked up by typing adoupdate, update. The development version of merlin in Stata, can be installed by typing:

net install merlin, from("https://www.mjcrowther.co.uk/code/merlin/")


and then run adoupdate as normal to pick up any updates. I may also change where the development version is hosted, so check back here every so often.

$~$

### merlin in R

The R version is currently in development, which began in September 2017. I aim to have a development version available soon.

• Tutorials in R
• Version history of merlin in R

$~$