My understanding is that it's exactly that. The modellers have to make some assumptions in order to be able to manage a simulation. You could take it to almost infinite levels of complexity, but at a certain point your model has to have some constraints, otherwise you can't make it manageable.
Since this is beyond my technical area of expertise, it might be more helpful if we, in follow-up to this session, perhaps provided a document outlining the methodology used in this—