Comment on "Renormalization group approach to connect discrete- and continuous-time descriptions of Gaussian processes".

F. Ferretti et al. [Phys. Rev. E 105, 044133 (2022)PREHBM2470-004510.1103/PhysRevE.105.044133] show that time discretization of linear Gaussian continuous-time stochastic processes are either first-order Markov processes or non-Markovian ones. Specializing to ARMA(2,1) processes, they propose a general redundantly parametrized form for a stochastic differential equation giving rise to this dynamics as well as a candidate nonredundant parametrization. However, the latter does not give rise to the full range of possible dynamics allowed by the former. I propose an alternative nonredundant parametrization which does.