Superstatistics is a relatively new proposal to explain the success of non-Boltzmann-Gibbs probability distributions in out-of-equilibrium systems and non-extensive systems. It amounts to a marginalization of the inverse temperature parameter β= 1/k BT over the joint distribution P (x, β| S). In this work we derive a fluctuation–dissipation theorem for Superstatistics which relates the expectations of the superstatistical model and the “pure” MaxEnt model, as well as present an inversion formula for the determination of P (β| S) from P (x| S) using MaxEnt.