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

Download:

Current browse context:

stat.ME

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

Statistics > Methodology

Title: Rollage: Efficient Rolling Average Algorithm to Estimate ARMA Models for Big Time Series Data

Abstract: We develop a new method to estimate an ARMA model in the presence of big time series data. Using the concept of a rolling average, we develop a new efficient algorithm, called Rollage, to estimate the order of an AR model and subsequently fit the model. When used in conjunction with an existing methodology, specifically Durbin's algorithm, we show that our proposed method can be used as a criterion to optimally fit ARMA models. Empirical results on large-scale synthetic time series data support the theoretical results and reveal the efficacy of this new approach, especially when compared to existing methodology.
Subjects: Methodology (stat.ME); Computation (stat.CO)
MSC classes: 62M10, 62R07
Cite as: arXiv:2103.09175 [stat.ME]
  (or arXiv:2103.09175v1 [stat.ME] for this version)

Submission history

From: Ali Eshragh [view email]
[v1] Tue, 16 Mar 2021 16:23:41 GMT (2180kb,D)

Link back to: arXiv, form interface, contact.