The intrinsic predictability of ecological time series and its potential to guide forecasting
(2019) In Ecological Monographs 89(2).- Abstract
Successfully predicting the future states of systems that are complex, stochastic, and potentially chaotic is a major challenge. Model forecasting error (FE) is the usual measure of success; however model predictions provide no insights into the potential for improvement. In short, the realized predictability of a specific model is uninformative about whether the system is inherently predictable or whether the chosen model is a poor match for the system and our observations thereof. Ideally, model proficiency would be judged with respect to the systems’ intrinsic predictability, the highest achievable predictability given the degree to which system dynamics are the result of deterministic vs. stochastic processes. Intrinsic... (More)
Successfully predicting the future states of systems that are complex, stochastic, and potentially chaotic is a major challenge. Model forecasting error (FE) is the usual measure of success; however model predictions provide no insights into the potential for improvement. In short, the realized predictability of a specific model is uninformative about whether the system is inherently predictable or whether the chosen model is a poor match for the system and our observations thereof. Ideally, model proficiency would be judged with respect to the systems’ intrinsic predictability, the highest achievable predictability given the degree to which system dynamics are the result of deterministic vs. stochastic processes. Intrinsic predictability may be quantified with permutation entropy (PE), a model-free, information-theoretic measure of the complexity of a time series. By means of simulations, we show that a correlation exists between estimated PE and FE and show how stochasticity, process error, and chaotic dynamics affect the relationship. This relationship is verified for a data set of 461 empirical ecological time series. We show how deviations from the expected PE–FE relationship are related to covariates of data quality and the nonlinearity of ecological dynamics. These results demonstrate a theoretically grounded basis for a model-free evaluation of a system's intrinsic predictability. Identifying the gap between the intrinsic and realized predictability of time series will enable researchers to understand whether forecasting proficiency is limited by the quality and quantity of their data or the ability of the chosen forecasting model to explain the data. Intrinsic predictability also provides a model-free baseline of forecasting proficiency against which modeling efforts can be evaluated.
(Less)
- author
- publishing date
- 2019-05-01
- type
- Contribution to journal
- publication status
- published
- keywords
- empirical dynamic modelling, forecasting, information theory, permutation entropy, population dynamics, time series analysis
- in
- Ecological Monographs
- volume
- 89
- issue
- 2
- article number
- e01359
- pages
- 17 pages
- publisher
- Ecological Society of America
- external identifiers
-
- scopus:85062595036
- ISSN
- 0012-9615
- DOI
- 10.1002/ecm.1359
- language
- English
- LU publication?
- no
- id
- 250bf5f0-8fc0-4cfe-981e-3b934d585681
- date added to LUP
- 2020-10-01 16:53:30
- date last changed
- 2022-04-19 01:00:33
@article{250bf5f0-8fc0-4cfe-981e-3b934d585681,
abstract = {{<p>Successfully predicting the future states of systems that are complex, stochastic, and potentially chaotic is a major challenge. Model forecasting error (FE) is the usual measure of success; however model predictions provide no insights into the potential for improvement. In short, the realized predictability of a specific model is uninformative about whether the system is inherently predictable or whether the chosen model is a poor match for the system and our observations thereof. Ideally, model proficiency would be judged with respect to the systems’ intrinsic predictability, the highest achievable predictability given the degree to which system dynamics are the result of deterministic vs. stochastic processes. Intrinsic predictability may be quantified with permutation entropy (PE), a model-free, information-theoretic measure of the complexity of a time series. By means of simulations, we show that a correlation exists between estimated PE and FE and show how stochasticity, process error, and chaotic dynamics affect the relationship. This relationship is verified for a data set of 461 empirical ecological time series. We show how deviations from the expected PE–FE relationship are related to covariates of data quality and the nonlinearity of ecological dynamics. These results demonstrate a theoretically grounded basis for a model-free evaluation of a system's intrinsic predictability. Identifying the gap between the intrinsic and realized predictability of time series will enable researchers to understand whether forecasting proficiency is limited by the quality and quantity of their data or the ability of the chosen forecasting model to explain the data. Intrinsic predictability also provides a model-free baseline of forecasting proficiency against which modeling efforts can be evaluated.</p>}},
author = {{Pennekamp, Frank and Iles, Alison C. and Garland, Joshua and Brennan, Georgina and Brose, Ulrich and Gaedke, Ursula and Jacob, Ute and Kratina, Pavel and Matthews, Blake and Munch, Stephan and Novak, Mark and Palamara, Gian Marco and Rall, Björn C. and Rosenbaum, Benjamin and Tabi, Andrea and Ward, Colette and Williams, Richard and Ye, Hao and Petchey, Owen L.}},
issn = {{0012-9615}},
keywords = {{empirical dynamic modelling; forecasting; information theory; permutation entropy; population dynamics; time series analysis}},
language = {{eng}},
month = {{05}},
number = {{2}},
publisher = {{Ecological Society of America}},
series = {{Ecological Monographs}},
title = {{The intrinsic predictability of ecological time series and its potential to guide forecasting}},
url = {{http://dx.doi.org/10.1002/ecm.1359}},
doi = {{10.1002/ecm.1359}},
volume = {{89}},
year = {{2019}},
}