Singular Spectrum Analysis of Univariate Time Series

Guillaume Mercère ()
Additional contact information
Guillaume Mercère: Université de Poitiers

Chapter Chapter 2 in Data Driven Model Learning for Engineers, 2023, pp 9-30 from Springer

Abstract: Abstract Before diving into statistical or numerical optimization considerations for characterizing time series parsimoniously and effectively, it is interesting to introduce a first technique, called the singular spectrum analysis (SSA), which yields the main components of a time series by resorting to applied linear algebra tools only. As shown in this chapter: The goal of SSA is to decompose a time series into a small number of interpretable components such as a trend, an oscillatory pattern, and noise. The SSA class of solutions can be used without making any statistical assumptions on the time series to be decomposed. The basic SSA solution assumes the existence of an unknown linear recurrent equation satisfied by the time series to be analyzed. The basic SSA algorithm is made of four steps and requires Hankel matrices and SVD only. SSA can be used for denoising a time series or, in a complementary way, for signal subspace extraction.

Date: 2023
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-031-31636-4_2

Ordering information: This item can be ordered from
http://www.springer.com/9783031316364

DOI: 10.1007/978-3-031-31636-4_2

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().