Quantifying model uncertainty for the observed non-Gaussian data by the Hellinger distance

DOI10.1016/J.CNSNS.2021.105720zbMATH Open1461.62166arXiv2011.11170OpenAlexW3109227607WikidataQ112880671 ScholiaQ112880671MaRDI QIDQ2656071FDOQ2656071

Authors: Yayun Zheng, Fang Yang, Jinqiao Duan, J. Kurths 

Publication date: 10 March 2021

Published in: Communications in Nonlinear Science and Numerical Simulation (Search for Journal in Brave)

Abstract: Mathematical models for complex systems under random fluctuations often certain uncertain parameters. However, quantifying model uncertainty for a stochastic differential equation with an $alpha$-stable L'evy process is still lacking. Here, we propose an approach to infer all the uncertain non-Gaussian parameters and other system parameters by minimizing the Hellinger distance over the parameter space. The Hellinger distance measures the similarity between an empirical probability density of non-Gaussian observations and a solution (as a probability density) of the associated nonlocal Fokker-Planck equation. Numerical experiments verify that our method is feasible for estimating single and multiple parameters. Meanwhile, we find an optimal estimation interval of the estimated parameters. This method is beneficial for extracting governing dynamical system models under non-Gaussian fluctuations, as in the study of abrupt climate changes in the Dansgaard-Oeschger events.

Full work available at URL: https://arxiv.org/abs/2011.11170

Recommendations

• Quantifying non-gaussianity via the Hellinger distance
• Quantifying model uncertainties in the space of probability measures
• Quantification of model uncertainty from data
• Uncertainty Quantification in Case of Imperfect Models: A Non‐Bayesian Approach
• Quantifying uncertainty for predictions with model error in non-Gaussian systems with intermittency
• Uncertainty quantification using the nearest neighbor Gaussian process
• Modeling Uncertainty and Nonlinearity by Probabilistic Metric Spaces
• On the quantification of model uncertainty: a Bayesian perspective

zbMATH Keywords

Hellinger distanceprobability densityparameters estimationnon-Gaussian observations

Mathematics Subject Classification ID

Point estimation (62F10) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Applications of statistics to environmental and related topics (62P12) Climate science and climate modeling (86A08) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Dynamical systems in fluid mechanics, oceanography and meteorology (37N10) Meteorology and atmospheric physics (86A10)

Cites Work

• Simulation and inference for stochastic differential equations. With R examples.
• Approximate self-weighted LAD estimation of discretely observed ergodic Ornstein-Uhlenbeck processes
• Title not available (Why is that?)
• Kernel density estimation via diffusion
• A simple general approach to inference about the tail of a distribution
• Parameter estimation in stochastic differential equations.
• Lévy Processes and Stochastic Calculus
• Minimum Hellinger distance estimates for parametric models
• Nonparametric estimation of diffusions: a differential equations approach
• An introduction to stochastic dynamics
• Least squares estimator for Ornstein-Uhlenbeck processes driven by \(\alpha \)-stable motions
• Densities for SDEs driven by degenerate \(\alpha\)-stable processes
• Parameter estimation for Ornstein-Uhlenbeck processes driven by \(\alpha\)-stable Lévy motions
• Quasi-likelihood analysis for the stochastic differential equation with jumps
• Analysis of data sets of stochastic systems
• Statistical estimation of multivariate Ornstein-Uhlenbeck processes and applications to co-integration
• Regularity of density for SDEs driven by degenerate Lévy noises
• Quantifying Model Uncertainties in Complex Systems
• Stochastic flows for Lévy processes with Hölder drifts
• Mathematics and climate
• Quantifying model uncertainty in dynamical systems driven by non-Gaussian Lévy stable noise with observations on mean exit time or escape probability
• Fokker-Planck equations for stochastic dynamical systems with symmetric Lévy motions
• Numerical algorithm for the space-time fractional Fokker-Planck system with two internal states
• Variational estimation of the drift for stochastic differential equations from the empirical density
• Generalized moment estimators for \(\alpha\)-stable Ornstein-Uhlenbeck motions from discrete observations
• Computational scheme for solving nonlinear fractional stochastic differential equations with delay
• Governing equations for probability densities of Marcus stochastic differential equations with Lévy noise
• An integro quadratic spline-based scheme for solving nonlinear fractional stochastic differential equations with constant time delay

Cited In (3)

• Quantifying non-gaussianity via the Hellinger distance
• Nonlinear dynamics of discontinuous uncertain oscillators with unilateral constraints
• Quantifying model uncertainty in dynamical systems driven by non-Gaussian Lévy stable noise with observations on mean exit time or escape probability

Uses Software

• sde

This page was built for publication: Quantifying model uncertainty for the observed non-Gaussian data by the Hellinger distance

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2656071)