Noise and nonlinearity in measles epidemics

Combining mechanistic and statistical approaches to population modeling

S. P. Ellner, B. A. Bailey, G. V. Bobashev, Andrew Ronald Gallant, B. T. Grenfell, D. W. Nychka

Research output: Contribution to journalArticle

105 Citations (Scopus)

Abstract

We present and evaluate an approach to analyzing population dynamics data using semimechanistic models. These models incorporate reliable information on population structure and underlying dynamic mechanisms but use nonparametric surface-fitting methods to avoid unsupported assumptions about the precise form of rate equations. Using historical data on measles epidemics as a case study, we show how this approach can lead to better forecasts, better characterizations of the dynamics, and a better understanding of the factors causing complex population dynamics relative to either mechanistic models or purely descriptive statistical time-series models. The semimechanistic models are found to have better forecasting accuracy than either of the model types used in previous analyses when tested on data not used to fit the models. The dynamics are characterized as being both nonlinear and noisy, and the global dynamics are clustered very tightly near the border of stability (dominant Lyapunov exponent λ ≃ 0). However, locally in state space the dynamics oscillate between strong short-term stability and strong short-term chaos (i.e., between negative and positive local Lyapunov exponents). There is statistically significant evidence for short-term chaos in all data sets examined. Thus the nonlinearity in these systems is characterized by the variance over state space in local measures of chaos versus stability rather than a single summary measure of the over-all dynamics as either chaotic or nonchaotic.

Original languageEnglish (US)
Pages (from-to)425-440
Number of pages16
JournalAmerican Naturalist
Volume151
Issue number5
DOIs
StatePublished - May 16 1998

Fingerprint

measles
population modeling
nonlinearity
chaotic dynamics
population dynamics
mechanistic models
time series analysis
population structure
case studies
time series

All Science Journal Classification (ASJC) codes

  • Ecology, Evolution, Behavior and Systematics

Cite this

Ellner, S. P. ; Bailey, B. A. ; Bobashev, G. V. ; Gallant, Andrew Ronald ; Grenfell, B. T. ; Nychka, D. W. / Noise and nonlinearity in measles epidemics : Combining mechanistic and statistical approaches to population modeling. In: American Naturalist. 1998 ; Vol. 151, No. 5. pp. 425-440.
@article{9212c914448e4103b76085311dd6cffd,
title = "Noise and nonlinearity in measles epidemics: Combining mechanistic and statistical approaches to population modeling",
abstract = "We present and evaluate an approach to analyzing population dynamics data using semimechanistic models. These models incorporate reliable information on population structure and underlying dynamic mechanisms but use nonparametric surface-fitting methods to avoid unsupported assumptions about the precise form of rate equations. Using historical data on measles epidemics as a case study, we show how this approach can lead to better forecasts, better characterizations of the dynamics, and a better understanding of the factors causing complex population dynamics relative to either mechanistic models or purely descriptive statistical time-series models. The semimechanistic models are found to have better forecasting accuracy than either of the model types used in previous analyses when tested on data not used to fit the models. The dynamics are characterized as being both nonlinear and noisy, and the global dynamics are clustered very tightly near the border of stability (dominant Lyapunov exponent λ ≃ 0). However, locally in state space the dynamics oscillate between strong short-term stability and strong short-term chaos (i.e., between negative and positive local Lyapunov exponents). There is statistically significant evidence for short-term chaos in all data sets examined. Thus the nonlinearity in these systems is characterized by the variance over state space in local measures of chaos versus stability rather than a single summary measure of the over-all dynamics as either chaotic or nonchaotic.",
author = "Ellner, {S. P.} and Bailey, {B. A.} and Bobashev, {G. V.} and Gallant, {Andrew Ronald} and Grenfell, {B. T.} and Nychka, {D. W.}",
year = "1998",
month = "5",
day = "16",
doi = "10.1086/286130",
language = "English (US)",
volume = "151",
pages = "425--440",
journal = "American Naturalist",
issn = "0003-0147",
publisher = "University of Chicago",
number = "5",

}

Noise and nonlinearity in measles epidemics : Combining mechanistic and statistical approaches to population modeling. / Ellner, S. P.; Bailey, B. A.; Bobashev, G. V.; Gallant, Andrew Ronald; Grenfell, B. T.; Nychka, D. W.

In: American Naturalist, Vol. 151, No. 5, 16.05.1998, p. 425-440.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Noise and nonlinearity in measles epidemics

T2 - Combining mechanistic and statistical approaches to population modeling

AU - Ellner, S. P.

AU - Bailey, B. A.

AU - Bobashev, G. V.

AU - Gallant, Andrew Ronald

AU - Grenfell, B. T.

AU - Nychka, D. W.

PY - 1998/5/16

Y1 - 1998/5/16

N2 - We present and evaluate an approach to analyzing population dynamics data using semimechanistic models. These models incorporate reliable information on population structure and underlying dynamic mechanisms but use nonparametric surface-fitting methods to avoid unsupported assumptions about the precise form of rate equations. Using historical data on measles epidemics as a case study, we show how this approach can lead to better forecasts, better characterizations of the dynamics, and a better understanding of the factors causing complex population dynamics relative to either mechanistic models or purely descriptive statistical time-series models. The semimechanistic models are found to have better forecasting accuracy than either of the model types used in previous analyses when tested on data not used to fit the models. The dynamics are characterized as being both nonlinear and noisy, and the global dynamics are clustered very tightly near the border of stability (dominant Lyapunov exponent λ ≃ 0). However, locally in state space the dynamics oscillate between strong short-term stability and strong short-term chaos (i.e., between negative and positive local Lyapunov exponents). There is statistically significant evidence for short-term chaos in all data sets examined. Thus the nonlinearity in these systems is characterized by the variance over state space in local measures of chaos versus stability rather than a single summary measure of the over-all dynamics as either chaotic or nonchaotic.

AB - We present and evaluate an approach to analyzing population dynamics data using semimechanistic models. These models incorporate reliable information on population structure and underlying dynamic mechanisms but use nonparametric surface-fitting methods to avoid unsupported assumptions about the precise form of rate equations. Using historical data on measles epidemics as a case study, we show how this approach can lead to better forecasts, better characterizations of the dynamics, and a better understanding of the factors causing complex population dynamics relative to either mechanistic models or purely descriptive statistical time-series models. The semimechanistic models are found to have better forecasting accuracy than either of the model types used in previous analyses when tested on data not used to fit the models. The dynamics are characterized as being both nonlinear and noisy, and the global dynamics are clustered very tightly near the border of stability (dominant Lyapunov exponent λ ≃ 0). However, locally in state space the dynamics oscillate between strong short-term stability and strong short-term chaos (i.e., between negative and positive local Lyapunov exponents). There is statistically significant evidence for short-term chaos in all data sets examined. Thus the nonlinearity in these systems is characterized by the variance over state space in local measures of chaos versus stability rather than a single summary measure of the over-all dynamics as either chaotic or nonchaotic.

UR - http://www.scopus.com/inward/record.url?scp=0031895048&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031895048&partnerID=8YFLogxK

U2 - 10.1086/286130

DO - 10.1086/286130

M3 - Article

VL - 151

SP - 425

EP - 440

JO - American Naturalist

JF - American Naturalist

SN - 0003-0147

IS - 5

ER -