Summary: This paper examines the problem of probabilistic characterization of nonlinear systems driven by normal and Poissonian white noise. By means of classical nonlinear transformation, the stochastic differential equation driven by external input is transformed into a parametric-type stochastic differential equation. Such equations are commonly handled with Itô-type stochastic differential equations, and Itô’s rule is used to find the response statistics. Here a different approach is proposed, which mainly consists in transforming the Fokker-Planck equation for the original system driven by external input, in the transformed probability density function of the new state variable. It will be shown that the Wong-Zakay or Stratonovich corrective terms and the hierarchy of correction terms in the case of Poissonian white noise arise in a natural way.