Motivated Metamodels: Synthesis of Cause-effect Reasoning and Statistical Metamodeling

A metamodel approximates the behavior of a more complex model. A common and superficially attractive way to develop a metamodel is to generate large-model data and use off-the-shelf statistical methods without attempting to understand the model’s internal workings. This report describes research illuminating why it can be important to improve the quality of metamodels by using even modest phenomenological knowledge to help structure them. The work helps to understand multiresolution, multiperspective modeling.