Outlier detection for stationary time series

被引：40

作者：

Choy, K ^{[1
]}

机构：

[1] Komazawa Univ, Tokyo, Japan

来源：

JOURNAL OF STATISTICAL PLANNING AND INFERENCE | 2001年 / 99卷 / 02期

关键词：

additive outlier (AO); innovations outlier (IO); maximum likelihood estimation (MLE); spectrum-based outlier detection algorithm (SODA); spectral density; stationary time series; Whittle-type estimation;

D O I：

10.1016/S0378-3758(01)00081-7

中图分类号：

O21 [概率论与数理统计]; C8 [统计学];

学科分类号：

020208 [统计学]; 070103 [概率论与数理统计]; 0714 [统计学];

摘要：

Linear models are considered for outliers and their effects in stationary time series. Instead of using the exact maximum likelihood or the least squares methods, the Whittle-type (see Whittle, Ark. Math. 2 (1953) 423-434) estimators are derived for these models and an iterative procedure consisting of estimation, detection, removing cycles to specify the observed outlier-contaminated time series is proposed. (C) 2001 Elsevier Science B.V. All rights reserved.

引用

页码：111 / 127

页数：17

共 23 条

[1]

OUTLIER DETECTION AND TIME-SERIES MODELING [J].

ABRAHAM, B ;

CHUANG, A .

TECHNOMETRICS, 1989, 31 (02) :241-248

[2]

ABRAHAM B, 1979, BIOMETRIKA, V66, P229, DOI 10.1093/biomet/66.2.229

[3]

[Anonymous], 1981, Time series data analysis and theory, DOI 10.1201/b15288-24

[4]

[Anonymous], 1994, Time Series Analysis: Univariate and Multivariate Analysis

[5]

INTERVENTION ANALYSIS WITH APPLICATIONS TO ECONOMIC AND ENVIRONMENTAL PROBLEMS [J].

BOX, GEP ;

TIAO, GC .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1975, 70 (349) :70-79

[6]

BRUCE AG, 1989, J ROY STAT SOC B MET, V51, P363

[7]

Burr I.W., 1976, STAT QUALITY CONTROL

[8]

CHANG I, 1983, TECHNOMETRICS

[9]

Cho S., 1993, J KOREAN STAT SOC, V22, P283

[10]

DELUNA X, 2001, IN PRESS J COMPUT GR

← 1 2 3 →