Semiparametric forecasting and filtering: correcting low-dimensional model error in parametric models

Abstract

Semiparametric forecasting and filtering are introduced as a method of addressing model errors arising from unresolved physical phenomena. While traditional parametric models are able to learn high-dimensional systems from small data sets, their rigid parametric structure makes them vulnerable to model error. On the other hand, nonparametric models have a very flexible structure, but they suffer from the curse-of-dimensionality and are not practical for high-dimensional systems. The semiparametric approach loosens the structure of a parametric model by fitting a data-driven nonparametric model for the parameters. Given a parametric dynamical model and a noisy data set of historical observations, an adaptive Kalman filter is used to extract a time-series of the parameter values. A nonparametric forecasting model for the parameters is built by projecting the discrete shift map onto a data-driven basis of smooth functions. Existing techniques for filtering and forecasting algorithms extend naturally to the semiparametric model which can effectively compensate for model error, with forecasting skill approaching that of the perfect model. Semiparametric forecasting and filtering are a generalization of statistical semiparametric models to time-dependent distributions evolving under dynamical systems.

References in corpus (3)

• Nonparametric forecasting of low-dimensional dynamical systems
• Adaptive Error Covariances Estimation Methods for Ensemble Kalman Filters
• Forecasting Turbulent Modes with Nonparametric Diffusion Models: Learning from noisy data

Cited by in corpus (10)

• Hybrid modeling and prediction of dynamical systems
• Machine Learning for Prediction with Missing Dynamics
• Kalman-Takens filtering in the presence of dynamical noise
• Forecasting Turbulent Modes with Nonparametric Diffusion Models: Learning from noisy data
• Manifold Learning with Contracting Observers for Data-driven Time-series Analysis
• Modeling of Missing Dynamical Systems: Deriving Parametric Models using a Nonparametric Framework
• Quantum mechanics and data assimilation
• Diffusion forecasting model with basis functions from QR decomposition
• Analog Forecasting with Dynamics-Adapted Kernels
• Dispersion based Recurrent Neural Network Model for Methane Monitoring in Albertan Tailings Ponds