2. Approximation methods

The reduction of high-dimensional linear dynamic models is not a new topic, and many techniques have been developed to address the problem. The aim of this section is to present some of these approaches. The aim is not to draw up an exhaustive list, which would be too long, but rather to highlight those approaches which, according to the authors, are the most interesting, either because of their popularity, or because of the possibilities they offer. Furthermore, to preserve the didactic character intended for this article, the emphasis is placed on the main principles underlying the various methods rather than on mathematical demonstrations and technical details, which readers can find in the references mentioned.

Considering the model-based approximation problem stated in section