Once the model is formulated and formalized, some rigorous model testing is in order. The major steps that are usually assumed are called sensitivity analysis, calibration, validation, and verification. While the first two can be well formalized and are quite standard in any modeling effort, the validation and verification stages are designed to assess the level of 'truth' that the model delivers, and therefore tend to be more vague and controversial. There has been a good deal of discussion about what a good model is, and whether

Figure 1 Analysis of sensitivity in a simple population model.


Figure 1 Analysis of sensitivity in a simple population model.

it is feasible at all to claim that the model is true in any sense. One can argue that for an open system, where conditions constantly change, it is not even possible to design a model that would represent reality, since the reality is constantly changing with additional factors brought in all the time. The model then can only represent the situation that it has been designed for and is very much limited by the conditions and factors that were included.

Nevertheless some model testing is definitely in order and some models are still better than others. In spite of much leeway in the definitions of what a good model is, the model testing and analysis is an important stage, which can tell us much about the system, even if it does not really tell us how 'true' the model is.

Was this article helpful?

0 0
Project Earth Conservation

Project Earth Conservation

Get All The Support And Guidance You Need To Be A Success At Helping Save The Earth. This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Recycle to Create a Better Future for Our Children.

Get My Free Ebook

Post a comment