An algorithm based on singular spectrum analysis for change-point detection

被引：144

作者：

Moskvina, V ^{[1
]}

Zhigljavsky, A ^{[1
]}

机构：

[1] Univ Wales Coll Cardiff, Sch Math, Cardiff CF24 4YH, S Glam, Wales

来源：

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION | 2003年 / 32卷 / 02期

关键词：

change-point detection; singular-spectrum analysis; sequential algorithm; singular value decomposition;

D O I：

10.1081/SAC-120017494

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 ; 070103 ; 0714 ;

摘要：

This paper is devoted to application of the singular-spectrum analysis to sequential detection of changes in time series. An algorithm of change-point detection in time series, based on sequential application of the singular-spectrum analysis is developed and studied. The algorithm is applied to different data sets and extensively studied numerically. For specific models, several numerical approximations to the error probabilities and the power function of the algorithm are obtained. Numerical comparisons with other methods are given.

引用

页码：319 / 352

页数：34

共 13 条

[1]

Bakhache B, 2000, INT J ADAPT CONTROL, V14, P683, DOI 10.1002/1099-1115(200011)14:7<683::AID-ACS616>3.0.CO

[2]

2-Z

[3]

Basseville M, 1993, DETECTION ABRUPT CHA

[4]

BOX GEP, 1970, TIME SERIES ANAL FOR

[5] EXTRACTING QUALITATIVE DYNAMICS FROM EXPERIMENTAL-DATA [J].

BROOMHEAD, DS ;

KING, GP .

PHYSICA D, 1986, 20 (2-3) :217-236

[6]

Danilov DL, 1997, Principal Components of Time Series: The 'Caterpillar' Method

[7]

ELSNER JB, 1996, SINGULAR SPECTRUM AN

[8]

GOLJANDINA NE, 2001, ANAL TIME SERIES STR

[9]

MOSKINA V, 2001, THESIS CARDIFF U CAR

[10]

MOSKVINA V, 2000, UNPUB J TIME SERIES

← 1 2 →