the AGRITALIM model generates a whole series of simulativeoutput indicators which can be analyzed using the RS approach, which is a type of nonparametric “supervised machine learning” method. This method can be used to study the downstream correlative links of the simulations, in order to analyze – in the different scenarios proposed – which drivers were most relevant in determining specific outcome variables of the simulation model.
The RS methodology computes farm-specific responsiveness scores using an iterated random-coefficient regression approach.