Efficiency of the Girsanov transformation approach for parametric sensitivity analysis of stochastic chemical kinetics

DOI10.1137/140998111zbMATH Open1355.92046arXiv1412.1005OpenAlexW2963273107MaRDI QIDQ3179330FDOQ3179330

Ting Wang, Muruhan Rathinam

Publication date: 21 December 2016

Published in: SIAM/ASA Journal on Uncertainty Quantification (Search for Journal in Brave)

Abstract: Most common Monte Carlo methods for sensitivity analysis of stochastic reaction networks are the finite difference (FD), the Girsanov transformation (GT) and the regularized pathwise derivative (RPD) methods. It has been numerically observed in the literature, that the biased FD and RPD methods tend to have lower variance than the unbiased GT method and that centering the GT method (CGT) reduces its variance. We provide a theoretical justification for these observations in terms of system size asymptotic analysis under what is known as the classical scaling. Our analysis applies to GT, CGT and FD, and shows that the standard deviations of their estimators when normalized by the actual sensitivity, scale as $mathcalO(N^{1/2}),mathcalO(1)$ and $mathcalO(N^{-1/2})$ respectively, as system size $Noinfty$. In the case of the FD methods, the $Noinfty$ asymptotics are obtained keeping the finite difference perturbation $h$ fixed. Our numerical examples verify that our order estimates are sharp and that the variance of the RPD method scales similarly to the FD methods. We combine our large $N$ asymptotics with previously known small $h$ asymptotics to obtain the best choice of $h$ in terms of $N$, and estimate the number $N_s$ of simulations required to achieve a prescribed relative $mathcal{L}_2$ error $delta$. This shows that $N_s$ depends on $delta$ and $N$ as $delt{a}^{-2-fracgamm{a}_{2}gamm{a}_1}{N}^{-1},delt{a}^{-2}$ and $Ndelt{a}^{-2}$, for FD, CGT and GT respectively. Here $gamm{a}_1>0,gamm{a}_2>0$ depend on the type of FD method used.

Full work available at URL: https://arxiv.org/abs/1412.1005

Recommendations

• On the validity of the Girsanov transformation method for sensitivity analysis of stochastic chemical reaction networks
• Efficient stochastic sensitivity analysis of discrete event systems
• An efficient finite difference method for parameter sensitivities of continuous time Markov chains
• Unbiased estimation of parameter sensitivities for stochastic chemical reaction networks
• Variance of finite difference methods for reaction networks with non-Lipschitz rate functions

zbMATH Keywords

variance analysisMonte Carlo methodsasymptotic analysisfinite difference methodGirsanov transformationstochastic chemical kineticsstochastic equationsparametric sensitivity

Mathematics Subject Classification ID

Monte Carlo methods (65C05) Computational methods for stochastic equations (aspects of stochastic analysis) (60H35) Kinetics in biochemical problems (pharmacokinetics, enzyme kinetics, etc.) (92C45) Classical flows, reactions, etc. in chemistry (92E20)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Stochastic simulation: Algorithms and analysis
• Weak limit theorems for stochastic integrals and stochastic differential equations
• Asymptotic analysis of multiscale approximations to reaction networks
• Point processes and queues. Martingale dynamics
• Separation of time-scales and model reduction for stochastic reaction networks
• Title not available (Why is that?)
• Proofs of the martingale FCLT
• Complexity of Multilevel Monte Carlo Tau-Leaping
• Title not available (Why is that?)
• Title not available (Why is that?)
• Reversible-equivalent-monomolecular tau: a leaping method for ``small number and stiff stochastic chemical systems
• Moment growth bounds on continuous time Markov processes on non-negative integer lattices
• An Efficient Finite Difference Method for Parameter Sensitivities of Continuous Time Markov Chains
• Efficient stochastic sensitivity analysis of discrete event systems
• Unbiased Estimation of Parameter Sensitivities for Stochastic Chemical Reaction Networks
• Sensitivity analysis for stochastic chemical reaction networks with multiple time-scales
• Homogenization of cadlag processes

Cited In (6)

• Steady-State Sensitivity Analysis of Continuous Time Markov Chains
• Convergence of the likelihood ratio method for linear response of non-equilibrium stationary states
• On the validity of the Girsanov transformation method for sensitivity analysis of stochastic chemical reaction networks
• Efficient stochastic sensitivity analysis of discrete event systems
• Multiscale Global Sensitivity Analysis for Stochastic Chemical Systems
• Estimation of Parameter Sensitivities for Stochastic Reaction Networks Using Tau-Leap Simulations

This page was built for publication: Efficiency of the Girsanov transformation approach for parametric sensitivity analysis of stochastic chemical kinetics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3179330)