Reprint of: Fitting population growth models in the presence of measurement and detection error

Trevor J. Hefley, Andrew J. Tyre, Erin E. Blankenship

Research output: Contribution to journalArticle

Abstract

Population time series data from field studies are complex and statistical analysis requires models that describe nonlinear population dynamics and observational errors. State-space formulations of stochastic population growth models have been used to account for measurement error caused by the data collection process. Parameter estimation, inference, and prediction are all sensitive to measurement error. The observational process may also result in detection errors and if unaccounted for will result in biased parameter estimates. We developed an N-mixture state-space modeling framework to estimate and correct for errors in detection while estimating population model parameters. We tested our methods using simulated data sets and compared the results to those obtained with state-space models when detection is perfect and when detection is ignored. Our N-mixture state-space model yielded parameter estimates of similar quality to a state-space model when detection is perfect. Our results show that ignoring detection errors can lead to biased parameter estimates including an overestimated growth rate, underestimated equilibrium population size and estimated population state that is misleading. We recommend that researchers consider the possibility of detection errors when collecting and analyzing population time series data.

Original languageEnglish (US)
Pages (from-to)461-467
Number of pages7
JournalEcological Modelling
Volume359
DOIs
StatePublished - Sep 10 2017

Fingerprint

population growth
time series
detection
population size
population dynamics
statistical analysis
parameter
prediction
modeling

Keywords

  • Calibrated Bayes
  • Detection error
  • N-mixture model
  • Observation error
  • State-space model
  • Theta logistic population growth model

ASJC Scopus subject areas

  • Ecological Modeling

Cite this

Reprint of : Fitting population growth models in the presence of measurement and detection error. / Hefley, Trevor J.; Tyre, Andrew J.; Blankenship, Erin E.

In: Ecological Modelling, Vol. 359, 10.09.2017, p. 461-467.

Research output: Contribution to journalArticle

@article{396708a5cffe4572854cc41fb007211e,
title = "Reprint of: Fitting population growth models in the presence of measurement and detection error",
abstract = "Population time series data from field studies are complex and statistical analysis requires models that describe nonlinear population dynamics and observational errors. State-space formulations of stochastic population growth models have been used to account for measurement error caused by the data collection process. Parameter estimation, inference, and prediction are all sensitive to measurement error. The observational process may also result in detection errors and if unaccounted for will result in biased parameter estimates. We developed an N-mixture state-space modeling framework to estimate and correct for errors in detection while estimating population model parameters. We tested our methods using simulated data sets and compared the results to those obtained with state-space models when detection is perfect and when detection is ignored. Our N-mixture state-space model yielded parameter estimates of similar quality to a state-space model when detection is perfect. Our results show that ignoring detection errors can lead to biased parameter estimates including an overestimated growth rate, underestimated equilibrium population size and estimated population state that is misleading. We recommend that researchers consider the possibility of detection errors when collecting and analyzing population time series data.",
keywords = "Calibrated Bayes, Detection error, N-mixture model, Observation error, State-space model, Theta logistic population growth model",
author = "Hefley, {Trevor J.} and Tyre, {Andrew J.} and Blankenship, {Erin E.}",
year = "2017",
month = "9",
day = "10",
doi = "10.1016/j.ecolmodel.2013.10.021",
language = "English (US)",
volume = "359",
pages = "461--467",
journal = "Ecological Modelling",
issn = "0304-3800",
publisher = "Elsevier",

}

TY - JOUR

T1 - Reprint of

T2 - Fitting population growth models in the presence of measurement and detection error

AU - Hefley, Trevor J.

AU - Tyre, Andrew J.

AU - Blankenship, Erin E.

PY - 2017/9/10

Y1 - 2017/9/10

N2 - Population time series data from field studies are complex and statistical analysis requires models that describe nonlinear population dynamics and observational errors. State-space formulations of stochastic population growth models have been used to account for measurement error caused by the data collection process. Parameter estimation, inference, and prediction are all sensitive to measurement error. The observational process may also result in detection errors and if unaccounted for will result in biased parameter estimates. We developed an N-mixture state-space modeling framework to estimate and correct for errors in detection while estimating population model parameters. We tested our methods using simulated data sets and compared the results to those obtained with state-space models when detection is perfect and when detection is ignored. Our N-mixture state-space model yielded parameter estimates of similar quality to a state-space model when detection is perfect. Our results show that ignoring detection errors can lead to biased parameter estimates including an overestimated growth rate, underestimated equilibrium population size and estimated population state that is misleading. We recommend that researchers consider the possibility of detection errors when collecting and analyzing population time series data.

AB - Population time series data from field studies are complex and statistical analysis requires models that describe nonlinear population dynamics and observational errors. State-space formulations of stochastic population growth models have been used to account for measurement error caused by the data collection process. Parameter estimation, inference, and prediction are all sensitive to measurement error. The observational process may also result in detection errors and if unaccounted for will result in biased parameter estimates. We developed an N-mixture state-space modeling framework to estimate and correct for errors in detection while estimating population model parameters. We tested our methods using simulated data sets and compared the results to those obtained with state-space models when detection is perfect and when detection is ignored. Our N-mixture state-space model yielded parameter estimates of similar quality to a state-space model when detection is perfect. Our results show that ignoring detection errors can lead to biased parameter estimates including an overestimated growth rate, underestimated equilibrium population size and estimated population state that is misleading. We recommend that researchers consider the possibility of detection errors when collecting and analyzing population time series data.

KW - Calibrated Bayes

KW - Detection error

KW - N-mixture model

KW - Observation error

KW - State-space model

KW - Theta logistic population growth model

UR - http://www.scopus.com/inward/record.url?scp=85045168024&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045168024&partnerID=8YFLogxK

U2 - 10.1016/j.ecolmodel.2013.10.021

DO - 10.1016/j.ecolmodel.2013.10.021

M3 - Article

AN - SCOPUS:85045168024

VL - 359

SP - 461

EP - 467

JO - Ecological Modelling

JF - Ecological Modelling

SN - 0304-3800

ER -