EUVIMEDEUVIMED entdecken
Crossref

Deep echo state networks with uncertainty quantification for spatio‐temporal forecasting

Patrick L. McDermott, Christopher K. Wikle

Environmetrics · 2019-05 · 30 (3)

DOI: 10.1002/env.2553 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Environmetrics, online erschienen am 18.12.2018. Verfasst von Patrick L. McDermott und 1 weiteren Autor:innen. Die Arbeit wurde bisher 76-mal zitiert (Crossref) und nennt 58 Literaturangaben. Gefördert von: National Science Foundation, U.S. Census Bureau.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Long‐lead forecasting for spatio‐temporal systems can entail complex nonlinear dynamics that are difficult to specify a priori. Current statistical methodologies for modeling these processes are often highly parameterized and, thus, challenging to implement from a computational perspective. One potential parsimonious solution to this problem is a method from the dynamical systems and engineering literature referred to as an echo state network (ESN). ESN models use reservoir computing to efficiently compute recurrent neural network forecasts. Moreover, multilevel (deep) hierarchical models have recently been shown to be successful at predicting high‐dimensional complex nonlinear processes, particularly those with multiple spatial and temporal scales of variability (such as those we often find in spatio‐temporal environmental data). Here, we introduce a deep ensemble ESN (D‐EESN) model. Despite the incorporation of a deep structure, the presented model is computationally efficient. We present two versions of this model for spatio‐temporal processes that produce forecasts and associated measures of uncertainty. The first approach utilizes a bootstrap ensemble framework, and the second is developed within a hierarchical Bayesian framework (BD‐EESN). This more general hierarchical Bayesian framework naturally accommodates non‐Gaussian data types and multiple levels of uncertainties. The methodology is first applied to a data set simulated from a novel non‐Gaussian multiscale Lorenz‐96 dynamical system simulation model and, then, to a long‐lead United States (U.S.) soil moisture forecasting application. Across both applications, the proposed methodology improves upon existing methods in terms of both forecast accuracy and quantifying uncertainty.

Förderung

  • National Science Foundation
  • U.S. Census Bureau

Zitieren

APA

McDermott, P. L., & Wikle, C. K. (2019). Deep echo state networks with uncertainty quantification for spatio‐temporal forecasting. Environmetrics, 30(3). https://doi.org/10.1002/env.2553

Vancouver

McDermott PL, Wikle CK. Deep echo state networks with uncertainty quantification for spatio‐temporal forecasting. Environmetrics. 2019;30(3). doi:10.1002/env.2553

BibTeX
@article{McDermott2019,
  author = {McDermott, Patrick L. and Wikle, Christopher K.},
  title = {Deep echo state networks with uncertainty quantification for spatio‐temporal forecasting},
  journal = {Environmetrics},
  year = {2019},
  volume = {30},
  number = {3},
  doi = {10.1002/env.2553}
}
Kostenlos starten →