MCMC Methods for Parameter Inference in Structurally Nonidentifiable Models
For researchers in systems biology, control theory, and public health who need to perform Bayesian inference on structurally nonidentifiable models, this work provides practical MCMC algorithms that mitigate poor mixing and slow convergence.
The paper develops two MCMC methods for parameter inference in ODE models with structural non-identifiability, achieving improved sampling efficiency and convergence over standard MCMC methods.
We consider the problem of parameter inference for ordinary differential equation (ODE) models with structural non-identifiability. Such models arise in a wide range of scientific fields, including control theory, systems biology, and public health. Structural non-identifiability occurs when distinct parameter values provide identical model outputs, resulting in lower-dimensional manifolds of observationally equivalent solutions in the parameter space. This poses challenges for Bayesian inference and Markov chain Monte Carlo (MCMC) methods, often leading to poor mixing and slow convergence. We develop two MCMC methods that use information from structural identifiability analysis. The first, Identifiability-Aware Geometric MCMC, constructs proposals that move within and between non-identifiable manifolds. The second, Identifiability-Aware Pseudo-Marginal MCMC, performs inference on the space of identifiable parameter combinations and reconstructs full parameter values. We show that both methods target the correct posterior distribution and are ergodic under standard conditions. Numerical examples demonstrate improved sampling efficiency and convergence compared with standard MCMC methods.