Maximum entropy in scale-invariant processes with a 1/$f^{\alpha}$ power spectrum: the effect of white noise anisotropy

Large value fluctuations are modeled by a system of nonlinear stochastic equations describing the interacting phase transitions. Under the action of anisotropic white noise, random processes are formed with the 1/$f^{\alpha}$ dependence of the power spectra on frequency at values of the exponent from 0.7 to 1.7. It is shown that fluctuations with 1/$f^{\alpha}$ power spectra in the studied range of changes correspond to the entropy maximum, which indicates the stability of processes with 1/$f^{\alpha}$ power spectra at different values of the exponent $\alpha$.