Signal identication in singular spectrum analysis

Atikur R. Khan, D. S. Poskitt

Research output: Contribution to journalArticleResearchpeer-review

Abstract

This paper provides an information theoretic analysis of the signal identification problem in singular spectrum analysis. We present a signal-plus-noise model based on the Karhunen-Loève expansion and use this model to motivate the construction of a minimum description length criterion that can be employed to identify the dimension (rank) of the signal component. We show that under very general regularity conditions the criterion will identify the true signal dimension with probability one as the sample size increases. A by-product of this analysis is a procedure for selecting a window length consistent with the Whitney embedding theorem. The upshot is a modeling strategy that results in a specification that yields a signal-noise reconstruction that minimises mean squared reconstruction error. Empirical results obtained using simulated and real world data series indicate that theoretical properties presented in the paper are reflected in observed behaviour, even in relatively small samples, and that the minimum description length modeling strategy provides the practitioner with an effective addition to the SSA tool box.
Original languageEnglish
Pages (from-to)71-98
Number of pages28
JournalAustralian and New Zealand Journal of Statistics
Volume58
Issue number1
DOIs
Publication statusPublished - Mar 2016

Cite this

@article{a66a5baed804449c93fcfe9e0591c973,
title = "Signal identication in singular spectrum analysis",
abstract = "This paper provides an information theoretic analysis of the signal identification problem in singular spectrum analysis. We present a signal-plus-noise model based on the Karhunen-Lo{\`e}ve expansion and use this model to motivate the construction of a minimum description length criterion that can be employed to identify the dimension (rank) of the signal component. We show that under very general regularity conditions the criterion will identify the true signal dimension with probability one as the sample size increases. A by-product of this analysis is a procedure for selecting a window length consistent with the Whitney embedding theorem. The upshot is a modeling strategy that results in a specification that yields a signal-noise reconstruction that minimises mean squared reconstruction error. Empirical results obtained using simulated and real world data series indicate that theoretical properties presented in the paper are reflected in observed behaviour, even in relatively small samples, and that the minimum description length modeling strategy provides the practitioner with an effective addition to the SSA tool box.",
author = "Khan, {Atikur R.} and Poskitt, {D. S.}",
year = "2016",
month = "3",
doi = "10.1111/anzs.12143",
language = "English",
volume = "58",
pages = "71--98",
journal = "Australian and New Zealand Journal of Statistics",
issn = "1369-1473",
publisher = "Wiley-Blackwell",
number = "1",

}

Signal identication in singular spectrum analysis. / Khan, Atikur R.; Poskitt, D. S.

In: Australian and New Zealand Journal of Statistics, Vol. 58, No. 1, 03.2016, p. 71-98.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Signal identication in singular spectrum analysis

AU - Khan, Atikur R.

AU - Poskitt, D. S.

PY - 2016/3

Y1 - 2016/3

N2 - This paper provides an information theoretic analysis of the signal identification problem in singular spectrum analysis. We present a signal-plus-noise model based on the Karhunen-Loève expansion and use this model to motivate the construction of a minimum description length criterion that can be employed to identify the dimension (rank) of the signal component. We show that under very general regularity conditions the criterion will identify the true signal dimension with probability one as the sample size increases. A by-product of this analysis is a procedure for selecting a window length consistent with the Whitney embedding theorem. The upshot is a modeling strategy that results in a specification that yields a signal-noise reconstruction that minimises mean squared reconstruction error. Empirical results obtained using simulated and real world data series indicate that theoretical properties presented in the paper are reflected in observed behaviour, even in relatively small samples, and that the minimum description length modeling strategy provides the practitioner with an effective addition to the SSA tool box.

AB - This paper provides an information theoretic analysis of the signal identification problem in singular spectrum analysis. We present a signal-plus-noise model based on the Karhunen-Loève expansion and use this model to motivate the construction of a minimum description length criterion that can be employed to identify the dimension (rank) of the signal component. We show that under very general regularity conditions the criterion will identify the true signal dimension with probability one as the sample size increases. A by-product of this analysis is a procedure for selecting a window length consistent with the Whitney embedding theorem. The upshot is a modeling strategy that results in a specification that yields a signal-noise reconstruction that minimises mean squared reconstruction error. Empirical results obtained using simulated and real world data series indicate that theoretical properties presented in the paper are reflected in observed behaviour, even in relatively small samples, and that the minimum description length modeling strategy provides the practitioner with an effective addition to the SSA tool box.

UR - https://www-scopus-com.ezproxy.lib.monash.edu.au/record/display.uri?eid=2-s2.0-84961692297&origin=resultslist&

U2 - 10.1111/anzs.12143

DO - 10.1111/anzs.12143

M3 - Article

VL - 58

SP - 71

EP - 98

JO - Australian and New Zealand Journal of Statistics

JF - Australian and New Zealand Journal of Statistics

SN - 1369-1473

IS - 1

ER -