A probabilistic approach to drift estimation from stochastic data

Abstract

Timeseries generated from a dynamical source can often be modeled as sample paths of a stochastic differential equation (SDE). The timeseries thus reflects the motion of a particle which flows along the direction provided by a drift / vector field, and is simultaneously scattered by the effect of white noise. The resulting motion can only be described as a random process instead of a solution curve. Due to the non-deterministic nature of this motion, the task of determining the drift from data is quite challenging, since the data does not directly represent the directional information of the flow. This paper describes an interpretation of a drift as a conditional expectation, which makes its estimation feasible via kernel-integral methods. In addition, some techniques are proposed to overcome the challenge of dimensionality if the SDE's carry some structure enabling sparsity. The technique is shown to be convergent, consistent and permits a wide choice of kernels.

References in corpus (17)

• Delay-coordinate maps and the spectra of Koopman operators
• Learning physics-constrained subgrid-scale closures in the small-data regime for stable and accurate LES
• Koopman spectra in reproducing kernel Hilbert spaces
• Reproducing kernel Hilbert space compactification of unitary evolution groups
• Approximate Bayes learning of stochastic differential equations
• Non-parametric Estimation of Stochastic Differential Equations with Sparse Gaussian Processes
• Stochastic geometric models with non-stationary spatial correlations in Lagrangian fluid flows
• Integral identity and measure estimates for stationary Fokker-Planck equations
• Out-of-sample generalizations for supervised manifold learning for classification
• One-Shot Learning of Stochastic Differential Equations with Data Adapted Kernels
• Manifold learning with bi-stochastic kernels
• Critical Nodes Identification in Complex Networks: A Survey
• Onsager's Conjecture for Subgrid Scale $\alpha$-Models of Turbulence
• Limits of Learning Dynamical Systems
• Constructing differential equations using only a scalar time-series about continuous time chaotic dynamics
• Data-driven discovery of quasiperiodically driven dynamics
• Conditional expectation using compactification operators