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

Download:

Current browse context:

math.ST

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 > Statistics Theory

Title: Generating Sparse Stochastic Processes Using Matched Splines

Abstract: We provide an algorithm to generate trajectories of sparse stochastic processes that are solutions of linear ordinary differential equations driven by L\'evy white noises. A recent paper showed that these processes are limits in law of generalized compound-Poisson processes. Based on this result, we derive an off-the-grid algorithm that generates arbitrarily close approximations of the target process. Our method relies on a B-spline representation of generalized compound-Poisson processes. We illustrate numerically the validity of our approach.
Subjects: Statistics Theory (math.ST)
DOI: 10.1109/TSP.2020.3011632
Cite as: arXiv:2008.03181 [math.ST]
  (or arXiv:2008.03181v1 [math.ST] for this version)

Submission history

From: Shayan Aziznejad [view email]
[v1] Fri, 7 Aug 2020 13:42:17 GMT (2202kb,D)

Link back to: arXiv, form interface, contact.