Home

pybisp : Bayesian inference for stochastic processes

This project aims to provide tools to estimate the parameters associated with the drift and diffusion coefficients of multi-dimensional diffusion processes,

dX = mu(X, theta) dt + sigma(X, theta) dW

where mu(X, theta) is the drift and sigma(X, theta) is the diffusion and dW is a Wiener increment.

Roadmap

1. Use pandas data frame to load time series data.
2. Write functions for exact inference for Wiener and Ornstein-Uhlenbeck processes.

Similar projects

• SDE in R
• Yuima Project in R
• Pyprocess in Python, to generate processes

Published resources

• Simulation and Inference of Stochastic Processes uses Yuima to illustrate the theory.