Efficient maximum likelihood parameterization of continuous-time Markov\n processes

2015/04/07 by Robert T. McGibbon, Vijay S. Pande, McGibbon, Robert T. +1 · 1 citation
Biochemistry, Genetics and Molecular Biology · Computer Science · #Biomolecules (q-bio.BM) #Chemical Physics (physics.chem-ph) #Data Analysis #FOS: Biological sciences #FOS: Physical sciences #Gaussian Processes and Bayesian Inference #Gene Regulatory Network Analysis #Statistics and Probability (physics.data-an)

paper · pdf · doi:10.48550/arxiv.1504.01804

openalex publication_date 2015/04/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Continuous-time Markov processes over finite state-spaces are widely used to\nmodel dynamical processes in many fields of natural and social science. Here,\nwe introduce an maximum likelihood estimator for constructing such models from\ndata observed at a finite time interval. This estimator is dramatically more\nefficient than prior approaches, enables the calculation of deterministic\nconfidence intervals in all model parameters, and can easily enforce important\nphysical constraints on the models such as detailed balance. We demonstrate and\ndiscuss the advantages of these models over existing discrete-time Markov\nmodels for the analysis of molecular dynamics simulations.\n

Cited by

• Efficient Inference for Coupled Hidden Markov Models in Continuous Time and Discrete Space

Related

• Efficient Continuous-Time Markov Chain Estimation
• Markov models of molecular kinetics: Generation and validation
• Markov state models revisited: Principles and algorithms for unbiased observables
• Continuous-Time Higher Order Markov Chains: Formulation and Parameter Estimation
• Unbiased estimation of equilibrium, rates, and committors from Markov state model analysis