MATHEMATICAL EVALUATION OF THE FIT OF A THEORY WITH EXPERIMENTAL-DATA

A statistical evaluation of the fit of a theoretical conception (model) with experimental data has been almost always based on the assumption that experiments are loaded with errors while theory is simply either right or wrong. In this paper an alternative model is suggested, assuming that the experiments are effectively exact as compared to the approximate theoretical description. Basic problems of this model are formulated. For the simplest problem, comparison of two theories on the basis of one data set, two general solutions are suggested: one using nonparametric statistics; one syntactical in character.