Bayesian parameter estimation in Ecolego using an adaptive Metropolis-Hastings-within-Gibbs algorithm

Abstract: Ecolego is scientific software that can be used to model diverse systems within fields such as radioecology and pharmacokinetics. The purpose of this research is to develop an algorithm for estimating the probability density functions of unknown parameters of Ecolego models. In order to do so, a general-purpose adaptive Metropolis-Hastings-within-Gibbs algorithm is developed and tested on some examples of Ecolego models. The algorithm works adequately on those models, which indicates that the algorithm could be integrated successfully into future versions of Ecolego.