Importance sampling for a robust and efficient multilevel Monte Carlo estimator for stochastic reaction networks
This repository includes Matlab codes and routines that were used in our manuscript entitled "Importance sampling for a robust and efficient multilevel Monte Carlo estimator for stochastic reaction networks" that can be found in https://link.springer.com/article/10.1007/s11222-020-09965-3 or in https://arxiv.org/abs/1911.06286. This work intends to design a more efficient and robust multilevel Monte Carlo (MLMC) estimator for stochastic biological systems and pure jumps. In particular, we address the issue of catastrophic coupling, explained in https://arxiv.org/abs/1911.06286.
The repository includes:
1- Folder “MLMC_without_IS_for_SRNs” includes all related code to MLMC, based on explicit tau-leap, without importance sampling
2- Folder “MLMC_with_IS_for_SRNs” includes all related code to MLMC, based on explicit tau-leap, with importance sampling.
3- Folder “MC_SSA_SRNs” includes all related code to Monte Carlo (MC) with an exact scheme SSA.
4- Folder “Produced_results” includes all obtained results in our manuscript.