Local Signal Detection for Categorical Time Series

David S. Stoffer ()
Additional contact information
David S. Stoffer: University of Pittsburgh

Chapter Chapter 23 in Research Papers in Statistical Inference for Time Series and Related Models, 2023, pp 519-538 from Springer

Abstract: Abstract Frequency domain signal detection for qualitative-valued time series was developed under the assumption of homogeneity using the concept of the spectral envelope. The technique was developed in relation to the optimal scaling of qualitative data. After reviewing some established results, we present a method for fitting a local spectral envelope to heterogeneous sequences based on a minimum description length criterion for choosing the best fitting model based on parsimony. In particular, we focus on the detection of breakpoints in long sequences. Because of the enormous size of the search space, optimization is accomplished using a genetic algorithm to effectively tackle the problem.

Keywords: Breakpoint detection; Genetic algorithm; Minimum description length; Nonhomogeneous processes; Qualitative time series; Scaling categorical data; Spectral envelope (search for similar items in EconPapers)
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:978-981-99-0803-5_23

Ordering information: This item can be ordered from
http://www.springer.com/9789819908035

DOI: 10.1007/978-981-99-0803-5_23

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 ().