Software to fit complex models using big data sets is needed to answer persistent and emerging questions in radiation epidemiology. The open-source R package Colossus was developed to meet this need. Colossus was designed to take advantage of the input and graphing flexibility of R scripts, employ multi-core systems to run analyses faster, and permit the straightforward addition of future capabilities. Incorporating methods to propagate covariate uncertainty into model parameter uncertainty is the next major focus area. Through guidance from NCRP Commentary 34, methods of analysing multiple realisations of exposure were implemented in Colossus. Frequentist model averaging and Monte Carlo maximum likelihood programs were added to Colossus to provide different methods of applying complex risk models to datasets with intricate exposure uncertainties.
Keywords: Million Person Study; Monte Carlo maximum likelihood; big data; frequentist model averaging; radiation epidemiology; uncertainty.
Creative Commons Attribution license.