Useful lies: Dynamics from data
The article deals with the construction of nonlinear dynamic models directly (i.e., without using first principles) from given time series. A primary goal is not a good fit but the model's ability to answer questions or to determine interesting quantities, e.g., reproducing dynamics, prediction horizon, Lyapunov exponents, cycles, bifurcations etc. The author discusses several approaches such as context tree reconstruction for problems with discrete data from a small alphabet and triangulation for nominally continuous data. Finally, open problems are indicated.NEWLINENEWLINEFor the entire collection see [Zbl 0958.00020].
- On selecting models for nonlinear time series
- Consistent nonlinear dynamics: identifying model inadequacy
- Data-driven reconstruction of nonlinear dynamics from sparse observation
- Automated reverse engineering of nonlinear dynamical systems
- scientific article; zbMATH DE number 4105912 (Why is no real title available?)
- NONPARAMETRIC MODELING AND SPATIOTEMPORAL DYNAMICAL SYSTEMS
- OPTIMAL NONLINEAR MODELS FROM EMPIRICAL TIME SERIES: AN APPLICATION TO CLIMATE
- scientific article; zbMATH DE number 934449 (Why is no real title available?)
- Adaptive, locally linear models of complex dynamics
- Nonlinear dynamical models from chaotic time series: methods and applications
This page was built for publication: Useful lies: Dynamics from data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2712144)