The Epistemic Structure of Economics Models and the Problem of Confirmation

Abstract

We propose that confirmation on economics models has a distinctive two-layered espistemic structure. That is, we analyze that the econometrician as the outside observer interferes with the very model she is observing. We further argue that, due to this structure and the impossibility of knowing the true probability distribution, the Market Selection Hypothesis (MSH) is not directly confirmable from data. We provide a formal proof of this result in the framework of a Bayesian confirmation model.