Analytic derivation of the statistical error in the Feynman-α method

The Feynman-α method is an often used method for analyzing the results of noise experiments in sub-critical systems near critical. In the study, we introduce new formulas, derived from first principals through the single energy point model stochastic transport equation, to evaluate the statistical error on the measured reactivity values. In particular, we demonstrate how the statistical error reduces dramatically as the system turns critical and the reactivity approaches zero.