/numapprox-is

Reliable inference in Bayesian models which require numerical approximations

Primary LanguageR

numapprox-is

An importance sampling framework for reliable and efficient inference in Bayesian models that require numerical approximations, such as solutions of implicitly defined functions. Ordinary differential equation (ODE) models are one example.

Experiments

Code for reproducing the experiments of the paper is in the experiments subdirectory. Running them requires the odemodeling R package, which was developed for these experiments. Experiments were run using version 0.2.0 of odemodeling.

References

Timonen, J., Siccha, N., Bales, B., Lähdesmäki, H., & Vehtari, A. (2023). An importance sampling approach for reliable and efficient inference in Bayesian ordinary differential equation models. Stat, 12(1), e614. https://doi.org/10.1002/sta4.614