Improvement on the innovational outlier detection procedure in a bilinear model

I. B. Mohamed, M. I. Ismail, M. S. Yahya, A. G. Hussin, N. Mohamed, Azami Zaharim, M. S. Zainol

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper considers the problem of outlier detection in bilinear time series data with special focus on BL(1,0,1,1) and BL(1,1,1,1) models. In the previous study, the formulations of effect of innovational outlier on the observations and residuals from the process had been developed and the corresponding least squares estimator of outlier effect had been derived. Consequently, an outlier detection procedure employing bootstrap-based procedure to estimate the variance of the estimator had been proposed. In this paper, we proposed to use the mean absolute deviance and trimmed mean formula to estimate the variance to improve the performances of the procedure. Via simulation, we showed that the procedure based on the trimmed mean formula has successfully improved the performance of the procedure.

Original languageEnglish
Pages (from-to)191-196
Number of pages6
JournalSains Malaysiana
Volume40
Issue number2
Publication statusPublished - Feb 2011

Fingerprint

Time series

Keywords

  • Bilinear
  • Bootstrap
  • Innovational outlier
  • Least squares method

ASJC Scopus subject areas

  • General

Cite this

Mohamed, I. B., Ismail, M. I., Yahya, M. S., Hussin, A. G., Mohamed, N., Zaharim, A., & Zainol, M. S. (2011). Improvement on the innovational outlier detection procedure in a bilinear model. Sains Malaysiana, 40(2), 191-196.

Improvement on the innovational outlier detection procedure in a bilinear model. / Mohamed, I. B.; Ismail, M. I.; Yahya, M. S.; Hussin, A. G.; Mohamed, N.; Zaharim, Azami; Zainol, M. S.

In: Sains Malaysiana, Vol. 40, No. 2, 02.2011, p. 191-196.

Research output: Contribution to journalArticle

Mohamed, IB, Ismail, MI, Yahya, MS, Hussin, AG, Mohamed, N, Zaharim, A & Zainol, MS 2011, 'Improvement on the innovational outlier detection procedure in a bilinear model', Sains Malaysiana, vol. 40, no. 2, pp. 191-196.
Mohamed IB, Ismail MI, Yahya MS, Hussin AG, Mohamed N, Zaharim A et al. Improvement on the innovational outlier detection procedure in a bilinear model. Sains Malaysiana. 2011 Feb;40(2):191-196.
Mohamed, I. B. ; Ismail, M. I. ; Yahya, M. S. ; Hussin, A. G. ; Mohamed, N. ; Zaharim, Azami ; Zainol, M. S. / Improvement on the innovational outlier detection procedure in a bilinear model. In: Sains Malaysiana. 2011 ; Vol. 40, No. 2. pp. 191-196.
@article{269d9219442549fb9d379404c08f5e41,
title = "Improvement on the innovational outlier detection procedure in a bilinear model",
abstract = "This paper considers the problem of outlier detection in bilinear time series data with special focus on BL(1,0,1,1) and BL(1,1,1,1) models. In the previous study, the formulations of effect of innovational outlier on the observations and residuals from the process had been developed and the corresponding least squares estimator of outlier effect had been derived. Consequently, an outlier detection procedure employing bootstrap-based procedure to estimate the variance of the estimator had been proposed. In this paper, we proposed to use the mean absolute deviance and trimmed mean formula to estimate the variance to improve the performances of the procedure. Via simulation, we showed that the procedure based on the trimmed mean formula has successfully improved the performance of the procedure.",
keywords = "Bilinear, Bootstrap, Innovational outlier, Least squares method",
author = "Mohamed, {I. B.} and Ismail, {M. I.} and Yahya, {M. S.} and Hussin, {A. G.} and N. Mohamed and Azami Zaharim and Zainol, {M. S.}",
year = "2011",
month = "2",
language = "English",
volume = "40",
pages = "191--196",
journal = "Sains Malaysiana",
issn = "0126-6039",
publisher = "Penerbit Universiti Kebangsaan Malaysia",
number = "2",

}

TY - JOUR

T1 - Improvement on the innovational outlier detection procedure in a bilinear model

AU - Mohamed, I. B.

AU - Ismail, M. I.

AU - Yahya, M. S.

AU - Hussin, A. G.

AU - Mohamed, N.

AU - Zaharim, Azami

AU - Zainol, M. S.

PY - 2011/2

Y1 - 2011/2

N2 - This paper considers the problem of outlier detection in bilinear time series data with special focus on BL(1,0,1,1) and BL(1,1,1,1) models. In the previous study, the formulations of effect of innovational outlier on the observations and residuals from the process had been developed and the corresponding least squares estimator of outlier effect had been derived. Consequently, an outlier detection procedure employing bootstrap-based procedure to estimate the variance of the estimator had been proposed. In this paper, we proposed to use the mean absolute deviance and trimmed mean formula to estimate the variance to improve the performances of the procedure. Via simulation, we showed that the procedure based on the trimmed mean formula has successfully improved the performance of the procedure.

AB - This paper considers the problem of outlier detection in bilinear time series data with special focus on BL(1,0,1,1) and BL(1,1,1,1) models. In the previous study, the formulations of effect of innovational outlier on the observations and residuals from the process had been developed and the corresponding least squares estimator of outlier effect had been derived. Consequently, an outlier detection procedure employing bootstrap-based procedure to estimate the variance of the estimator had been proposed. In this paper, we proposed to use the mean absolute deviance and trimmed mean formula to estimate the variance to improve the performances of the procedure. Via simulation, we showed that the procedure based on the trimmed mean formula has successfully improved the performance of the procedure.

KW - Bilinear

KW - Bootstrap

KW - Innovational outlier

KW - Least squares method

UR - http://www.scopus.com/inward/record.url?scp=79952685393&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952685393&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:79952685393

VL - 40

SP - 191

EP - 196

JO - Sains Malaysiana

JF - Sains Malaysiana

SN - 0126-6039

IS - 2

ER -