Abstract

A method for estimating parameters in dynamic stochastic (Markov Chain) models based on Kurtz's limit theory coupled with inverse problem methods developed for deterministic dynamical systems is proposed and illustrated in the context of disease dynamics. This methodology relies on finding an approximate large-population behavior of an appropriate scaled stochastic system. The approach leads to a deterministic approximation obtained as solutions of rate equations (ordinary differential equations) in terms of the large sample size average over sample paths or trajectories (limits of pure jump Markov processes). Using the resulting deterministic model, we select parameter subset combinations that can be estimated using an ordinary-least-squares (OLS) or generalized-least-squares (GLS) inverse problem formulation with a given data set. The selection is based on two criteria of the sensitivity matrix: the degree of sensitivity measured in the form of its condition number and the degree of uncertainty measured in the form of its parameter selection score. We illustrate the ideas with a stochastic model for the transmission of vancomycin-resistant enterococcus (VRE) in hospitals and VRE surveillance data from an oncology unit.

Original languageEnglish (US)
Pages (from-to)71-100
Number of pages30
JournalJournal of Biological Systems
Volume19
Issue number1
DOIs
StatePublished - Mar 2011

Fingerprint

Markov Chains
Markov Chain Model
Markov chain
Least-Squares Analysis
Inverse problems
Markov processes
Inverse Problem
vancomycin
Enterococcus
inverse problem
Markov Jump Processes
Oncology
Generalized Least Squares
Ordinary Least Squares
methodology
Methodology
least squares
Stochastic systems
Parameter Selection
Rate Equations

Keywords

  • Inverse Problems
  • Large Population Sample Path Approximations
  • Markov Chain Stochastic Models
  • Parameter Estimation
  • Parameter Selection

ASJC Scopus subject areas

  • Agricultural and Biological Sciences (miscellaneous)
  • Ecology
  • Applied Mathematics

Cite this

A deterministic methodology for estimation of parameters in dynamic markov chain models. / Ortiz, A. R.; Banks, H. T.; Castillo-Chavez, Carlos; Chowell, G.; Wang, X.

In: Journal of Biological Systems, Vol. 19, No. 1, 03.2011, p. 71-100.

Research output: Contribution to journalArticle

Ortiz, A. R. ; Banks, H. T. ; Castillo-Chavez, Carlos ; Chowell, G. ; Wang, X. / A deterministic methodology for estimation of parameters in dynamic markov chain models. In: Journal of Biological Systems. 2011 ; Vol. 19, No. 1. pp. 71-100.
@article{39294e1cf1b44e17a4f85b2a88e8236e,
title = "A deterministic methodology for estimation of parameters in dynamic markov chain models",
abstract = "A method for estimating parameters in dynamic stochastic (Markov Chain) models based on Kurtz's limit theory coupled with inverse problem methods developed for deterministic dynamical systems is proposed and illustrated in the context of disease dynamics. This methodology relies on finding an approximate large-population behavior of an appropriate scaled stochastic system. The approach leads to a deterministic approximation obtained as solutions of rate equations (ordinary differential equations) in terms of the large sample size average over sample paths or trajectories (limits of pure jump Markov processes). Using the resulting deterministic model, we select parameter subset combinations that can be estimated using an ordinary-least-squares (OLS) or generalized-least-squares (GLS) inverse problem formulation with a given data set. The selection is based on two criteria of the sensitivity matrix: the degree of sensitivity measured in the form of its condition number and the degree of uncertainty measured in the form of its parameter selection score. We illustrate the ideas with a stochastic model for the transmission of vancomycin-resistant enterococcus (VRE) in hospitals and VRE surveillance data from an oncology unit.",
keywords = "Inverse Problems, Large Population Sample Path Approximations, Markov Chain Stochastic Models, Parameter Estimation, Parameter Selection",
author = "Ortiz, {A. R.} and Banks, {H. T.} and Carlos Castillo-Chavez and G. Chowell and X. Wang",
year = "2011",
month = "3",
doi = "10.1142/S0218339011003798",
language = "English (US)",
volume = "19",
pages = "71--100",
journal = "Journal of Biological Systems",
issn = "0218-3390",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "1",

}

TY - JOUR

T1 - A deterministic methodology for estimation of parameters in dynamic markov chain models

AU - Ortiz, A. R.

AU - Banks, H. T.

AU - Castillo-Chavez, Carlos

AU - Chowell, G.

AU - Wang, X.

PY - 2011/3

Y1 - 2011/3

N2 - A method for estimating parameters in dynamic stochastic (Markov Chain) models based on Kurtz's limit theory coupled with inverse problem methods developed for deterministic dynamical systems is proposed and illustrated in the context of disease dynamics. This methodology relies on finding an approximate large-population behavior of an appropriate scaled stochastic system. The approach leads to a deterministic approximation obtained as solutions of rate equations (ordinary differential equations) in terms of the large sample size average over sample paths or trajectories (limits of pure jump Markov processes). Using the resulting deterministic model, we select parameter subset combinations that can be estimated using an ordinary-least-squares (OLS) or generalized-least-squares (GLS) inverse problem formulation with a given data set. The selection is based on two criteria of the sensitivity matrix: the degree of sensitivity measured in the form of its condition number and the degree of uncertainty measured in the form of its parameter selection score. We illustrate the ideas with a stochastic model for the transmission of vancomycin-resistant enterococcus (VRE) in hospitals and VRE surveillance data from an oncology unit.

AB - A method for estimating parameters in dynamic stochastic (Markov Chain) models based on Kurtz's limit theory coupled with inverse problem methods developed for deterministic dynamical systems is proposed and illustrated in the context of disease dynamics. This methodology relies on finding an approximate large-population behavior of an appropriate scaled stochastic system. The approach leads to a deterministic approximation obtained as solutions of rate equations (ordinary differential equations) in terms of the large sample size average over sample paths or trajectories (limits of pure jump Markov processes). Using the resulting deterministic model, we select parameter subset combinations that can be estimated using an ordinary-least-squares (OLS) or generalized-least-squares (GLS) inverse problem formulation with a given data set. The selection is based on two criteria of the sensitivity matrix: the degree of sensitivity measured in the form of its condition number and the degree of uncertainty measured in the form of its parameter selection score. We illustrate the ideas with a stochastic model for the transmission of vancomycin-resistant enterococcus (VRE) in hospitals and VRE surveillance data from an oncology unit.

KW - Inverse Problems

KW - Large Population Sample Path Approximations

KW - Markov Chain Stochastic Models

KW - Parameter Estimation

KW - Parameter Selection

UR - http://www.scopus.com/inward/record.url?scp=79953199801&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953199801&partnerID=8YFLogxK

U2 - 10.1142/S0218339011003798

DO - 10.1142/S0218339011003798

M3 - Article

VL - 19

SP - 71

EP - 100

JO - Journal of Biological Systems

JF - Journal of Biological Systems

SN - 0218-3390

IS - 1

ER -