Random coefficient regression in ASReml-R

C. Supakorn


Repeated measurements on individuals, often referred to as longitudinal data, is common in agricultural and human health research. Such data are recorded along a continuous scale, usually time. For example, data on individual cows’ milk production over a milking season or growth data on individual children as they age. Random coefficient regression can be used when we want to model means over time of repeated records, i.e. the average trajectory.

In random coefficient regression some parameters are treated as fixed effects to account for overall trends or effects, whilst others are treated as correlated random effects to allow for individual variation in the shape of trajectory and the correlation between the repeated measurements on the same individual.


ASReml-R, developed by VSNi, is a powerful statistical software tool specially designed for fitting mixed models, such as random coefficient regression, in the R environment. VSNi helps you understand more about mixed model analysis via video tutorials and other training material. Watch this free video to learn about random coefficient regression in ASReml-R!



In this video tutorial you will learn how to perform random coefficient regression in ASReml-R, as well as how to interpret the results. The video uses a nice example of longitudinal data, from an experiment to study the effect of three different drugs (control, thyroxine, and thiouracil) on the body weight of rats over time. The aim of the experiment was to compare the average weight profiles of the three drug treatments.