In this case, you may refer to Neel's 1946 phase theory. In general you may associate this fluctuation to magnetic domain wall dynamics, which is a quasi-periodic fluctuation. I am attaching an interesting paper.
If a system is at thermodynamic equilibrium, large fluctuations occur in a given property Z if the total entropy S(Z) of the system plus surroundings combined, or of the system alone if it is isolated, varies only very slowly with the value of that property. The probability of a fluctuation of Z away from its most probable value Zo corresponding to entropy S(Zo) at thermodynamic equilibrium is given by the inverted Boltzmann relation, P(Z)dZ = P(Zo)exp{-[S(Zo) - S(Z)]/k}dZ. If S(Z) decreases only very slowly from S(Zo) as Z varies from Zo, fluctuations in Z away from Zo will be large.