We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.NA

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Numerical Analysis

Title: Comparison of continuous and discrete-time data-based modeling for hypoelliptic systems

Abstract: We compare two approaches to the predictive modeling of dynamical systems from partial observations at discrete times. The first is continuous in time, where one uses data to infer a model in the form of stochastic differential equations, which are then discretized for numerical solution. The second is discrete in time, where one directly infers a discrete-time model in the form of a nonlinear autoregression moving average model. The comparison is performed in a special case where the observations are known to have been obtained from a hypoelliptic stochastic differential equation. We show that the discrete-time approach has better predictive skills, especially when the data are relatively sparse in time. We discuss open questions as well as the broader significance of the results.
Comments: 25 pages
Subjects: Numerical Analysis (math.NA)
MSC classes: 65C60, 62M09, 62M20
Journal reference: Commun. Appl. Math. Comput. Sci. 11 (2016) 187-216
DOI: 10.2140/camcos.2016.11.187
Cite as: arXiv:1605.02273 [math.NA]
  (or arXiv:1605.02273v2 [math.NA] for this version)

Submission history

From: Fei Lu [view email]
[v1] Sun, 8 May 2016 04:21:25 GMT (1445kb,D)
[v2] Wed, 7 Dec 2016 08:13:13 GMT (579kb,D)

Link back to: arXiv, form interface, contact.