---
_id: '10794'
abstract:
- lang: eng
text: Mathematical models are of fundamental importance in the understanding of
complex population dynamics. For instance, they can be used to predict the population
evolution starting from different initial conditions or to test how a system responds
to external perturbations. For this analysis to be meaningful in real applications,
however, it is of paramount importance to choose an appropriate model structure
and to infer the model parameters from measured data. While many parameter inference
methods are available for models based on deterministic ordinary differential
equations, the same does not hold for more detailed individual-based models. Here
we consider, in particular, stochastic models in which the time evolution of the
species abundances is described by a continuous-time Markov chain. These models
are governed by a master equation that is typically difficult to solve. Consequently,
traditional inference methods that rely on iterative evaluation of parameter likelihoods
are computationally intractable. The aim of this paper is to present recent advances
in parameter inference for continuous-time Markov chain models, based on a moment
closure approximation of the parameter likelihood, and to investigate how these
results can help in understanding, and ultimately controlling, complex systems
in ecology. Specifically, we illustrate through an agricultural pest case study
how parameters of a stochastic individual-based model can be identified from measured
data and how the resulting model can be used to solve an optimal control problem
in a stochastic setting. In particular, we show how the matter of determining
the optimal combination of two different pest control methods can be formulated
as a chance constrained optimization problem where the control action is modeled
as a state reset, leading to a hybrid system formulation.
acknowledgement: "The authors would like to acknowledge contributions from Baptiste
Mottet who performed preliminary analysis regarding parameter inference for the
considered case study in a student project (Mottet, 2014/2015).\r\nThe research
leading to these results has received funding from the People Programme (Marie Curie
Actions) of the European Union's Seventh Framework Programme (FP7/2007-2013) under
REA grant agreement No. [291734] and from SystemsX under the project SignalX."
article_number: '42'
article_processing_charge: No
article_type: original
author:
- first_name: Francesca
full_name: Parise, Francesca
last_name: Parise
- first_name: John
full_name: Lygeros, John
last_name: Lygeros
- first_name: Jakob
full_name: Ruess, Jakob
id: 4A245D00-F248-11E8-B48F-1D18A9856A87
last_name: Ruess
orcid: 0000-0003-1615-3282
citation:
ama: 'Parise F, Lygeros J, Ruess J. Bayesian inference for stochastic individual-based
models of ecological systems: a pest control simulation study. Frontiers in
Environmental Science. 2015;3. doi:10.3389/fenvs.2015.00042'
apa: 'Parise, F., Lygeros, J., & Ruess, J. (2015). Bayesian inference for stochastic
individual-based models of ecological systems: a pest control simulation study.
Frontiers in Environmental Science. Frontiers. https://doi.org/10.3389/fenvs.2015.00042'
chicago: 'Parise, Francesca, John Lygeros, and Jakob Ruess. “Bayesian Inference
for Stochastic Individual-Based Models of Ecological Systems: A Pest Control Simulation
Study.” Frontiers in Environmental Science. Frontiers, 2015. https://doi.org/10.3389/fenvs.2015.00042.'
ieee: 'F. Parise, J. Lygeros, and J. Ruess, “Bayesian inference for stochastic individual-based
models of ecological systems: a pest control simulation study,” Frontiers in
Environmental Science, vol. 3. Frontiers, 2015.'
ista: 'Parise F, Lygeros J, Ruess J. 2015. Bayesian inference for stochastic individual-based
models of ecological systems: a pest control simulation study. Frontiers in Environmental
Science. 3, 42.'
mla: 'Parise, Francesca, et al. “Bayesian Inference for Stochastic Individual-Based
Models of Ecological Systems: A Pest Control Simulation Study.” Frontiers in
Environmental Science, vol. 3, 42, Frontiers, 2015, doi:10.3389/fenvs.2015.00042.'
short: F. Parise, J. Lygeros, J. Ruess, Frontiers in Environmental Science 3 (2015).
date_created: 2022-02-25T11:42:25Z
date_published: 2015-06-10T00:00:00Z
date_updated: 2022-02-25T11:59:23Z
day: '10'
ddc:
- '000'
- '570'
department:
- _id: ToHe
- _id: GaTk
doi: 10.3389/fenvs.2015.00042
ec_funded: 1
file:
- access_level: open_access
checksum: 26c222487564e1be02a11d688d6f769d
content_type: application/pdf
creator: dernst
date_created: 2022-02-25T11:55:26Z
date_updated: 2022-02-25T11:55:26Z
file_id: '10795'
file_name: 2015_FrontiersEnvironmScience_Parise.pdf
file_size: 1371201
relation: main_file
success: 1
file_date_updated: 2022-02-25T11:55:26Z
has_accepted_license: '1'
intvolume: ' 3'
keyword:
- General Environmental Science
language:
- iso: eng
month: '06'
oa: 1
oa_version: Published Version
project:
- _id: 25681D80-B435-11E9-9278-68D0E5697425
call_identifier: FP7
grant_number: '291734'
name: International IST Postdoc Fellowship Programme
publication: Frontiers in Environmental Science
publication_identifier:
issn:
- 2296-665X
publication_status: published
publisher: Frontiers
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'Bayesian inference for stochastic individual-based models of ecological systems:
a pest control simulation study'
tmp:
image: /images/cc_by.png
legal_code_url: https://creativecommons.org/licenses/by/4.0/legalcode
name: Creative Commons Attribution 4.0 International Public License (CC-BY 4.0)
short: CC BY (4.0)
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 3
year: '2015'
...