Data-based inference of generators for markov jump processes using convex optimization

Daan Crommelin, Eric Vanden Eijnden

Research output: Contribution to journalArticle

Abstract

A variational approach to the estimation of generators for Markov jump processes from discretely sampled data is discussed and generalized. In this approach, one first calculates the spectrum of the discrete maximum likelihood estimator for the transition matrix consistent with the discrete data. Then the generator that best matches the spectrum is determined by solving a convex quadratic minimization problem with linear constraints (quadratic program). Here, we discuss the method in detail and position it in the context of maximum likelihood inference of generators from discretely sampled data. Furthermore, we show how the approach can be generalized to estimation from data sampled at nonconstant time intervals. Finally, we discuss numerical aspects of the algorithm for estimation of processes with high-dimensional state spaces. Numerical examples are presented throughout the paper.

Original languageEnglish (US)
Pages (from-to)1751-1778
Number of pages28
JournalMultiscale Modeling and Simulation
Volume7
Issue number4
DOIs
StatePublished - 2009

Fingerprint

Markov Jump Processes
Markov processes
Convex optimization
Process Optimization
Convex Optimization
inference
generators
Generator
Maximum likelihood
optimization
Likelihood Inference
Quadratic Program
Discrete Data
Transition Matrix
Variational Approach
Linear Constraints
estimators
Maximum Likelihood Estimator
Minimization Problem
Maximum Likelihood

Keywords

  • Estimation
  • Generator
  • Markov jump process
  • Quadratic program

ASJC Scopus subject areas

  • Modeling and Simulation
  • Chemistry(all)
  • Computer Science Applications
  • Ecological Modeling
  • Physics and Astronomy(all)

Cite this

Data-based inference of generators for markov jump processes using convex optimization. / Crommelin, Daan; Vanden Eijnden, Eric.

In: Multiscale Modeling and Simulation, Vol. 7, No. 4, 2009, p. 1751-1778.

Research output: Contribution to journalArticle

@article{4ba500fb79e94f0890e9f77407a34cae,
title = "Data-based inference of generators for markov jump processes using convex optimization",
abstract = "A variational approach to the estimation of generators for Markov jump processes from discretely sampled data is discussed and generalized. In this approach, one first calculates the spectrum of the discrete maximum likelihood estimator for the transition matrix consistent with the discrete data. Then the generator that best matches the spectrum is determined by solving a convex quadratic minimization problem with linear constraints (quadratic program). Here, we discuss the method in detail and position it in the context of maximum likelihood inference of generators from discretely sampled data. Furthermore, we show how the approach can be generalized to estimation from data sampled at nonconstant time intervals. Finally, we discuss numerical aspects of the algorithm for estimation of processes with high-dimensional state spaces. Numerical examples are presented throughout the paper.",
keywords = "Estimation, Generator, Markov jump process, Quadratic program",
author = "Daan Crommelin and {Vanden Eijnden}, Eric",
year = "2009",
doi = "10.1137/080735977",
language = "English (US)",
volume = "7",
pages = "1751--1778",
journal = "Multiscale Modeling and Simulation",
issn = "1540-3459",
publisher = "Society for Industrial and Applied Mathematics Publications",
number = "4",

}

TY - JOUR

T1 - Data-based inference of generators for markov jump processes using convex optimization

AU - Crommelin, Daan

AU - Vanden Eijnden, Eric

PY - 2009

Y1 - 2009

N2 - A variational approach to the estimation of generators for Markov jump processes from discretely sampled data is discussed and generalized. In this approach, one first calculates the spectrum of the discrete maximum likelihood estimator for the transition matrix consistent with the discrete data. Then the generator that best matches the spectrum is determined by solving a convex quadratic minimization problem with linear constraints (quadratic program). Here, we discuss the method in detail and position it in the context of maximum likelihood inference of generators from discretely sampled data. Furthermore, we show how the approach can be generalized to estimation from data sampled at nonconstant time intervals. Finally, we discuss numerical aspects of the algorithm for estimation of processes with high-dimensional state spaces. Numerical examples are presented throughout the paper.

AB - A variational approach to the estimation of generators for Markov jump processes from discretely sampled data is discussed and generalized. In this approach, one first calculates the spectrum of the discrete maximum likelihood estimator for the transition matrix consistent with the discrete data. Then the generator that best matches the spectrum is determined by solving a convex quadratic minimization problem with linear constraints (quadratic program). Here, we discuss the method in detail and position it in the context of maximum likelihood inference of generators from discretely sampled data. Furthermore, we show how the approach can be generalized to estimation from data sampled at nonconstant time intervals. Finally, we discuss numerical aspects of the algorithm for estimation of processes with high-dimensional state spaces. Numerical examples are presented throughout the paper.

KW - Estimation

KW - Generator

KW - Markov jump process

KW - Quadratic program

UR - http://www.scopus.com/inward/record.url?scp=79954600165&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79954600165&partnerID=8YFLogxK

U2 - 10.1137/080735977

DO - 10.1137/080735977

M3 - Article

AN - SCOPUS:79954600165

VL - 7

SP - 1751

EP - 1778

JO - Multiscale Modeling and Simulation

JF - Multiscale Modeling and Simulation

SN - 1540-3459

IS - 4

ER -