A Julia package that makes it easy to define diffusion processes and sample from their laws.
DiffusionDefinition.jl comprises of:
- convenient methods facilitating
- defining diffusion laws
- forward-sampling their trajectories
- computing functionals of sampled paths
- computing gradients of functionals of sampled paths with respect to diffusion parameters or with respect to the starting point of the trajectory
- A number of predefined diffusion processes that can be immediately loaded in and experimented on
It is designed to work efficiently in a setting of Bayesian inference for diffusion processes.
It is a part of a larger suite of packages JuliaDiffusionBayes—a collection of efficient and intuitive tools for performing Bayesian inference for diffusion processes.