Operator-theoretic framework for forecasting nonlinear time series with kernel analog techniques
From MaRDI portal
Publication:2125604
Abstract: Kernel analog forecasting (KAF), alternatively known as kernel principal component regression, is a kernel method used for nonparametric statistical forecasting of dynamically generated time series data. This paper synthesizes descriptions of kernel methods and Koopman operator theory in order to provide a single consistent account of KAF. The framework presented here illuminates the property of the KAF method that, under measure-preserving and ergodic dynamics, it consistently approximates the conditional expectation of observables that are acted upon by the Koopman operator of the dynamical system and are conditioned on the observed data at forecast initialization. More precisely, KAF yields optimal predictions, in the sense of minimal root mean square error with respect to the invariant measure, in the asymptotic limit of large data. The presented framework facilitates, moreover, the analysis of generalization error and quantification of uncertainty. Extensions of KAF to the construction of conditional variance and conditional probability functions, as well as to non-symmetric kernels, are also shown. Illustrations of various aspects of KAF are provided with applications to simple examples, namely a periodic flow on the circle and the chaotic Lorenz 63 system.
Recommendations
- Kernel-based prediction of non-Markovian time series
- Analog forecasting with dynamics-adapted kernels
- Kernel Analog Forecasting: Multiscale Test Problems
- Prediction of high-dimensional time series with exogenous variables using generalized Koopman operator framework in reproducing kernel Hilbert space
- Nonstationary time series forecasting using wavelets and kernel smoothing
- Koopman operator framework for time series modeling and analysis
- A Functional Wavelet–Kernel Approach for Time Series Prediction
- Identification of non-linear time series via kernels
Cites work
- scientific article; zbMATH DE number 3121911 (Why is no real title available?)
- scientific article; zbMATH DE number 3810550 (Why is no real title available?)
- scientific article; zbMATH DE number 3573677 (Why is no real title available?)
- scientific article; zbMATH DE number 1516956 (Why is no real title available?)
- scientific article; zbMATH DE number 1866304 (Why is no real title available?)
- scientific article; zbMATH DE number 3105296 (Why is no real title available?)
- 10.1162/15324430260185646
- A computational method to extract macroscopic variables and their dynamics in multiscale systems
- A data-driven approximation of the koopman operator: extending dynamic mode decomposition
- A topological delay embedding theorem for infinite-dimensional dynamical systems
- An introduction to the theory of reproducing kernel Hilbert spaces
- Analog forecasting with dynamics-adapted kernels
- Analysis of the 3DVAR filter for the partially observed Lorenz '63 model
- Applied Koopmanism
- Bi-stochastic kernels via asymmetric affinity functions
- Comparison of systems with complex behavior
- Comparisons of new nonlinear modeling techniques with applications to infant respiration.
- Consistency of spectral clustering
- Data-driven model reduction and transfer operator approximation
- Data-driven spectral analysis of the Koopman operator
- Data-driven spectral decomposition and forecasting of ergodic dynamical systems
- Delay-coordinate maps and the spectra of Koopman operators
- Deterministic Nonperiodic Flow
- Diffusion forecasting model with basis functions from QR-decomposition
- Diffusion maps
- Dynamic mode decomposition of numerical and experimental data
- Eigendecompositions of transfer operators in reproducing kernel Hilbert spaces
- Eigenvalue decay rates for positive integral operators
- Eigenvalue distribution of compact operators
- Eigenvalues of integral operators defined by smooth positive definite kernels
- Embedology
- Ergodic theory, dynamic mode decomposition, and computation of spectral properties of the Koopman operator
- Estimating Long-Term Behavior of Flows without Trajectory Integration: The Infinitesimal Generator Approach
- Extracting qualitative dynamics from experimental data
- Foundations of Modern Probability
- From graph to manifold Laplacian: the convergence rate
- Geometric harmonics: a novel tool for multiscale out-of-sample extension of empirical functions
- Hamiltonian Systems and Transformation in Hilbert Space
- Integral manifolds and inertial manifolds for dissipative partial differential equations
- Interpolation of scattered data: distance matrices and conditionally positive definite functions
- Kernel flows: from learning kernels from data into the abyss
- Kernel methods in machine learning
- Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
- Learning Theory
- Learning Theory
- Local kernels and the geometric structure of data
- Nonlinear Laplacian spectral analysis for time series with intermittency and low-frequency variability
- Nonlinear prediction of chaotic time series
- On Learning Vector-Valued Functions
- On learning with integral operators
- On the Approximation of Complicated Dynamical Behavior
- On the numerical approximation of the Perron-Frobenius and Koopman operator
- Operator theoretic aspects of ergodic theory
- Positive definiteness, reproducing kernel Hilbert spaces and beyond
- Singular spectrum analysis in nonlinear dynamics, with applications to paleoclimatic time series
- Spatiotemporal pattern extraction by spectral analysis of vector-valued observables
- Spectral analysis of nonlinear flows
- Spectral properties of dynamical systems, model reduction and decompositions
- Support Vector Machines
- The Lorenz attractor exists
- The Lorenz attractor is mixing
- Theoretical foundations of the potential function method in pattern recognition learning
- Theory of Reproducing Kernels
- Time-scale separation from diffusion-mapped delay coordinates
- Traceable integral kernels on countably generated measure spaces
- Turbulence, Coherent Structures, Dynamical Systems and Symmetry
- Universality, Characteristic Kernels and RKHS Embedding of Measures
- Variable bandwidth diffusion kernels
- What are SRB measures, and which dynamical systems have them?
Cited in
(23)- A framework for machine learning of model error in dynamical systems
- Learning dynamical systems from data: a simple cross-validation perspective. III: Irregularly-sampled time series
- Simplicity bias, algorithmic probability, and the random logistic map
- Learning to Forecast Dynamical Systems from Streaming Data
- Bridging algorithmic information theory and machine learning: a new approach to kernel learning
- Hausdorff metric based training of kernels to learn attractors with application to 133 chaotic dynamical systems
- Learning dynamical systems from data: a simple cross-validation perspective. V: Sparse kernel flows for 132 chaotic dynamical systems
- Kernel-based prediction of non-Markovian time series
- Solving PDEs on unknown manifolds with machine learning
- Kernel Analog Forecasting: Multiscale Test Problems
- Kernel methods for center manifold approximation and a weak data-based version of the center manifold theorem
- Learning dynamical systems from data: a simple cross-validation perspective. I: Parametric kernel flows
- Koopman operator framework for time series modeling and analysis
- One-shot learning of stochastic differential equations with data adapted kernels
- Reproducing kernel Hilbert space compactification of unitary evolution groups
- Learning dynamical systems from data: a simple cross-validation perspective. IV: Case with partial observations
- A note on microlocal kernel design for some slow-fast stochastic differential equations with critical transitions and application to EEG signals
- Error bounds of the invariant statistics in machine learning of ergodic Itô diffusions
- Learning Theory for Dynamical Systems
- Analog forecasting with dynamics-adapted kernels
- Nonlinear model reduction for slow-fast stochastic systems near unknown invariant manifolds
- Ensemble forecasts in reproducing kernel Hilbert space family
- Prediction of high-dimensional time series with exogenous variables using generalized Koopman operator framework in reproducing kernel Hilbert space
This page was built for publication: Operator-theoretic framework for forecasting nonlinear time series with kernel analog techniques
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2125604)
