Statistical Inference for Stochastic Processes Statistics and Probability
Guided simulation of conditioned chemical reaction networks
Abstract
Abstract Let X be a chemical reaction process, modeled as a multi-dimensional continuous-time jump process. Assume that at given times $$0< t_1< \cdots <t_n$$ 0 < t 1 < ⋯ < t n , linear combinations $$v_i = L_i X(t_i),\, i=1,\dots ,n$$ v i = L i X ( t i ) , i = 1 , ⋯ , n are observed for given matrices $$L_i$$ L i . We show how the process that is conditioned on hitting the states $$v_1,\dots , v_n$$ v 1 , ⋯ , v n is obtained by a change of measure on the law of the unconditioned process. This results in an algorithm for obtaining weighted samples from the conditioned process. Our results are illustrated by numerical simulations.