probnum.diffeq.perturbed.step.perturb_lognormal(rng, step, solver_order, noise_scale, size=())[source]

Perturb the step with log-normally distributed noise.

Proposed by Abdulle and Garegnani (2020) 1.

  • rng (Generator) – Random number generator

  • step (Real) – Unperturbed step propesed by the steprule

  • solver_order (Integral) – Order of the solver

  • noise_scale (Real) – Scales the perturbation

  • size (Union[Integral, Iterable[Integral], None]) – Number of perturbation samples to be drawn. Optional. Default is size=().



Abdulle, A. and Garegnani, G. Random time step probabilistic methods for uncertainty quantification in chaotic and geometric numerical integration. Statistics and Computing. 2020.

Return type

Union[float, ndarray]