SDE-based models
These models are a type of aggregated or compartmental model, which is described by a system of initial value problems (IVP) given by stochastic differential equations (SDE). In MEmilio, they are implemented as an ODE-based model with an additional function to compute the random noise, as can be seen here. Hence, for the most part, SDE models are used exactly like ODE-based models. They mostly differ in how they are simulated, see the Simulation section below. For everything else, check out the page on ODE-based model usage.
The class used for implementing SDE models is called StochasticModel. It is derived from a CompartmentalModel (or optionally a FlowModel) for the representation of the deterministic part of the model equations. Check out SDE model creation for more details.
Simulation
Once the model is set up, one can run a simple simulation from time t0 to tmax with an initial step size dt
using the mio::simulate_stochastic() function. This will run a simulation of type StochasticSimulation that
saves the sizes of each compartment over time.
The simulation uses an Euler-Maruyama scheme by default, so the step size does not change over time.
Flow information cannot be obtained even when the StochasticModel is defined using a FlowModel, as the integrator may need to rescale results with respect to compartments to avoid negative values.