Local Spectral Analysis of Qualitative Sequences via Minimum Description Length

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

A chapter in Robust and Multivariate Statistical Methods, 2023, pp 477-495 from Springer

Abstract: Abstract The idea of signal detection in the frequency domain for qualitative-valued time series was developed in Stoffer et al. (Biometrika 80(3):611–622, 1993) under the assumption of homogeneity. The tool is called the spectral envelope and is related to the concept of scaling qualitative data. After reviewing the basic ideas, we present a method for fitting a local spectral envelope to heterogeneous sequences based on a minimum description length (MDL) criterion for choosing the best fitting model based on parsimony. Inherent in the methodology is the detection of breakpoints in long sequences. Because the search space is immense, optimization is accomplished via a genetic algorithm (GA) to effectively tackle the problem. Numerical examples are given using sleep state data and DNA sequences.

Keywords: Breakpoint detection; Categorical time series; Genetic Algorithm; MDL; Nonstationary processes; 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:978-3-031-22687-8_22

Ordering information: This item can be ordered from
http://www.springer.com/9783031226878

DOI: 10.1007/978-3-031-22687-8_22

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 ().