EvoPER-An R package for applying evolutionary computation methods in the parameter estimation of individual-based models implemented in Repast

摘要

Individual-based models are complex and they normally have an elevated number of input parametersudwhich must be tuned in order to reproduce the experimental or observed data as accurately as possible.udHence one of the weakest points of such kind of models is the fact that rarely the modeler has the enoughudinformation about the correct values or even the acceptable range for the input parameters. Therefore,udseveral parameter combinations must be checked to find an acceptable set of input factors minimizing theuddeviations of simulated and observed data. In practice, most of the times, is computationally unfeasible toudtraverse the complete search space to check all parameter combination in order to find the best of them.udThat is precisely the kind of combinatorial problem suitable for evolutionary computation techniques.udIn this work we present the EvoPER, an R package for simplifying the parameter estimation usingudevolutionary computation techniques. The current version of EvoPER includes implementations of PSO,udSA and ACO algorithms for parameter estimation of models generated with the open source agent-basedudmodeling toolkit Repast.