rEDM: An R Package for Empirical Dynamic Modeling and Convergent Cross Mapping

Abstract

Empirical dynamic modeling (EDM) is an emerging non-parametric framework for modeling nonlinear dynamic systems. EDM is based on the mathematicaltheory of reconstructing attractor manifolds from time series data (Takens 1981). The rEDM package collects several EDM methods, including simplexprojection (Sugihara and May 1990), S-map (Sugihara 1994), multivariate embeddings (Dixon et al. 1999), convergent cross mapping (Sugihara et al.2012), and multiview embedding (Ye and Sugihara 2016). Here, we introduce the basic underlying theory, and describe the functionality of the rEDM, usingexamples from both model simulations and real data.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 05, 2019
Accession Number
AD1135399

Entities

People

  • Adam Clark
  • Ethan R Deyle
  • George Sugihara
  • Hao Ye

Organizations

  • Scripps Institution of Oceanography

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Atmospheric Sciences
  • Composite Materials
  • Contracts
  • Covariance
  • Cross Correlation
  • Data Science
  • Databases
  • Delphi Method
  • Department Of Defense
  • Dynamics
  • Ecology
  • Ecosystems
  • Embedding
  • Equations
  • Gaussian Processes
  • Information Processing
  • Information Science
  • Linear Systems
  • Nonlinear Analysis
  • Nonlinear Dynamics
  • Precipitation
  • Productivity
  • Sea Surface Temperature
  • Standards
  • Statistical Algorithms
  • Statistical Analysis
  • Statistics

Readers

  • Neural Network Machine Learning.
  • Wave Propagation and Nonlinear Chaotic Dynamics.