Towards Generic Simulation for Demanding Stochastic Processes

We outline and test a new methodology for genuine simulation of stochastic processes with any dependence structure marginal distribution. reproduce time generalized, symmetric or asymmetric, moving-average scheme. This implements linear filtering non-Gaussian white noise, the weights filter determined by analytical equations, in terms autocovariance process. approximate distribution process, irrespective its type, using number cumulants, which turn determine cumulants manner that can readily support generation random numbers from approximation, so it be applicable simulation. The method is as uses process interest directly, without transformation (e.g., normalization). illustrate synthetic real-world applications, either persistence antipersistence, distributions are bounded, thus making problem more demanding. These include bounded both sides, such uniform, below, exponential Pareto, possibly having discontinuity at origin (intermittence). All examples studied show satisfactory performance method.