ABSTRACT
In our consulting practice, we are sometimes asked if dynamic models are “right” and the answer to that question is a resounding “No.” All models are theories and theories, by definition, are abstractions from reality. Box and Deming were not being evasive; they were being literal. Any type of model, regardless of the methodology or statistical sophistication, is wrong from the beginning. This concept goes beyond accepted technical measures of model validity, such as R-squared or other “goodness of fit” metrics. It is a fundamental principle of the real world like gravity, death, and taxes. Models are intended to be simplifications, so beware of any model that is said to represent “truth.” Wise modeling practitioners leave such matters to the theologians.
