Managing Large Sets Of Models

In recent years computer hardware and software has improved to such an extent that it is common to fit large numbers of models, either to ensure the ”best” is found or to combine the results from all of them. However, standard software is usually designed for fitting single models, perhaps including some residual analysis and model refinement, but not for analysing many models. Fitting large numbers of models and summarising the results effectively has to be done mainly by hand. This paper describes the development of a software, Moret, which automatically records and stores the results of models produced in R, providing overviews of all models fitted and ready access to model criteria and estimates for metamodel analyses.