Detecting points of change in time series

Tep Sastri, Benito Flores, Juan B Valdes

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

A performance comparison study of six time-series change detection procedures via forecast-monitoring simulation is presented. Four of the procedures are due to Brown [1], Page [2], Box and Tiao [3] and Gardner [4]. The other two sequential detection schemes are developed in this paper; the first is based on Bagshaw and Johnson [5], while the second employs a moving-block estimator of innovations autocovariance. Eighty synthetic time series were generated, using an autoregressive model, an integrated-moving average model and a time-regression model having step changes in the parameters. The statistics being employed are based upon cumulative sum (cusum), squared cusum and discounted cusum of the innovations from the corresponding forecasting models. Results of this study indicate that the procedures which employ squared cusum and nondiscounted cusum of innovations yield smaller rates of false detection. The two best change-detection statistics of this study appear to possess good potential for industrial and business forecast monitoring applications.

Original languageEnglish (US)
Pages (from-to)271-293
Number of pages23
JournalComputers and Operations Research
Volume16
Issue number3
DOIs
StatePublished - 1989
Externally publishedYes

Fingerprint

Cumulative Sum
time series
Time series
Innovation
innovation
business forecast
forecast procedure
Change Detection
statistics
Statistics
Forecast
monitoring
performance comparison
Monitoring
Sequential Detection
Autocovariance
Moving Average Model
Integrated Model
Performance Comparison
Autoregressive Model

ASJC Scopus subject areas

  • Information Systems and Management
  • Management Science and Operations Research
  • Applied Mathematics
  • Modeling and Simulation
  • Transportation

Cite this

Detecting points of change in time series. / Sastri, Tep; Flores, Benito; Valdes, Juan B.

In: Computers and Operations Research, Vol. 16, No. 3, 1989, p. 271-293.

Research output: Contribution to journalArticle

Sastri, Tep ; Flores, Benito ; Valdes, Juan B. / Detecting points of change in time series. In: Computers and Operations Research. 1989 ; Vol. 16, No. 3. pp. 271-293.
@article{f9fc917863d24d1ebf1f13d24b6ada5e,
title = "Detecting points of change in time series",
abstract = "A performance comparison study of six time-series change detection procedures via forecast-monitoring simulation is presented. Four of the procedures are due to Brown [1], Page [2], Box and Tiao [3] and Gardner [4]. The other two sequential detection schemes are developed in this paper; the first is based on Bagshaw and Johnson [5], while the second employs a moving-block estimator of innovations autocovariance. Eighty synthetic time series were generated, using an autoregressive model, an integrated-moving average model and a time-regression model having step changes in the parameters. The statistics being employed are based upon cumulative sum (cusum), squared cusum and discounted cusum of the innovations from the corresponding forecasting models. Results of this study indicate that the procedures which employ squared cusum and nondiscounted cusum of innovations yield smaller rates of false detection. The two best change-detection statistics of this study appear to possess good potential for industrial and business forecast monitoring applications.",
author = "Tep Sastri and Benito Flores and Valdes, {Juan B}",
year = "1989",
doi = "10.1016/0305-0548(89)90027-0",
language = "English (US)",
volume = "16",
pages = "271--293",
journal = "Surveys in Operations Research and Management Science",
issn = "0305-0548",
publisher = "Elsevier Limited",
number = "3",

}

TY - JOUR

T1 - Detecting points of change in time series

AU - Sastri, Tep

AU - Flores, Benito

AU - Valdes, Juan B

PY - 1989

Y1 - 1989

N2 - A performance comparison study of six time-series change detection procedures via forecast-monitoring simulation is presented. Four of the procedures are due to Brown [1], Page [2], Box and Tiao [3] and Gardner [4]. The other two sequential detection schemes are developed in this paper; the first is based on Bagshaw and Johnson [5], while the second employs a moving-block estimator of innovations autocovariance. Eighty synthetic time series were generated, using an autoregressive model, an integrated-moving average model and a time-regression model having step changes in the parameters. The statistics being employed are based upon cumulative sum (cusum), squared cusum and discounted cusum of the innovations from the corresponding forecasting models. Results of this study indicate that the procedures which employ squared cusum and nondiscounted cusum of innovations yield smaller rates of false detection. The two best change-detection statistics of this study appear to possess good potential for industrial and business forecast monitoring applications.

AB - A performance comparison study of six time-series change detection procedures via forecast-monitoring simulation is presented. Four of the procedures are due to Brown [1], Page [2], Box and Tiao [3] and Gardner [4]. The other two sequential detection schemes are developed in this paper; the first is based on Bagshaw and Johnson [5], while the second employs a moving-block estimator of innovations autocovariance. Eighty synthetic time series were generated, using an autoregressive model, an integrated-moving average model and a time-regression model having step changes in the parameters. The statistics being employed are based upon cumulative sum (cusum), squared cusum and discounted cusum of the innovations from the corresponding forecasting models. Results of this study indicate that the procedures which employ squared cusum and nondiscounted cusum of innovations yield smaller rates of false detection. The two best change-detection statistics of this study appear to possess good potential for industrial and business forecast monitoring applications.

UR - http://www.scopus.com/inward/record.url?scp=0024944244&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024944244&partnerID=8YFLogxK

U2 - 10.1016/0305-0548(89)90027-0

DO - 10.1016/0305-0548(89)90027-0

M3 - Article

AN - SCOPUS:0024944244

VL - 16

SP - 271

EP - 293

JO - Surveys in Operations Research and Management Science

JF - Surveys in Operations Research and Management Science

SN - 0305-0548

IS - 3

ER -