mmider | mschauer | SebaGraz |
BridgeSDEInference.jl is a software package for Julia programming language in which we implement various inference algorithms for diffusion processes based on Guided proposals. It is hosted on github and can be installed using Julia’s built-in package manager:
using Pkg
Pkg.add("BridgeSDEInference")
We provide routines for running an MCMC sampler that infers posterior distribution over the unknown parameters θ governing the stochastic differential equation
dXt=bθ(t,Xt)dt+σθ(t,Xt)dWt,t∈[0,T],X0=x0,from discrete-time, partial and noisy observations of the process X:
Vti=LiXti+ξi,i=1,…,Nwhere ξi are iid Gaussian random variables and Li∈Rmi×d.
We also provide routines for inference from first passage time observations, as well as for mixed effect models (the last one is still in development).
Check out the documentation for more details.