Reality-Based Probability & Statistics: Solving the Evidential Crisis

Abstract

It is past time to abandon significance testing. In case there is any reluctance to embrace this decision, proofs against the validity of testing to make decisions
or to identify cause are given. In their place, models should be cast in their reality-based, predictive form. This form can be used for model selection, observable
predictions, or for explaining outcomes. Cause is the ultimate explanation; yet the understanding of cause in modeling is severely lacking. All models should undergo formal verification, where their predictions are tested against competitors and against reality.