Performances test statistics for single outlier detection in bilinear (1,1,1,1) models

Azami Zaharim, Ibrahim Mohamed, Ibrahim Ahmad, Shahrum Abdullah, Mohd. Zaidi Omar

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

An outlier detection procedure for BL(1,1,1,1) model is developed based on the maxima of the test statistics measuring the effects of IO, AO, TC and LC. A simulation study is carried out in order to investigate the sampling properties of the maxima of the outlier test statistics. It is associated with the sample size, the type of outlier and the coefficients chosen for BL(1,1,1,1). The results show that, in general, the performance of the detection procedure is good. The outlier detection procedure performs well in detecting AO for large value of ω̂AO. As for IO, the performance of outlier detection procedure is better for model with larger coefficient values. The outlier detection procedure is capable of detecting TC and LC, though the performance is affected if ω is large.

Original languageEnglish
Pages (from-to)1359-1364
Number of pages6
JournalWSEAS Transactions on Mathematics
Volume5
Issue number12
Publication statusPublished - Dec 2006

Fingerprint

Outlier Detection
Performance Test
Test Statistic
Statistics
Outlier
Sampling
Model
Coefficient
Sample Size
Outlier detection
Test statistic
Simulation Study
Outliers
Coefficients

Keywords

  • Models
  • Nonlinear least squares
  • Outlier detection procedure
  • Test statistics

ASJC Scopus subject areas

  • Mathematics (miscellaneous)
  • Computational Mathematics
  • Computer Science (miscellaneous)

Cite this

Performances test statistics for single outlier detection in bilinear (1,1,1,1) models. / Zaharim, Azami; Mohamed, Ibrahim; Ahmad, Ibrahim; Abdullah, Shahrum; Omar, Mohd. Zaidi.

In: WSEAS Transactions on Mathematics, Vol. 5, No. 12, 12.2006, p. 1359-1364.

Research output: Contribution to journalArticle

@article{75d534e484b240adb2659989ff053dd8,
title = "Performances test statistics for single outlier detection in bilinear (1,1,1,1) models",
abstract = "An outlier detection procedure for BL(1,1,1,1) model is developed based on the maxima of the test statistics measuring the effects of IO, AO, TC and LC. A simulation study is carried out in order to investigate the sampling properties of the maxima of the outlier test statistics. It is associated with the sample size, the type of outlier and the coefficients chosen for BL(1,1,1,1). The results show that, in general, the performance of the detection procedure is good. The outlier detection procedure performs well in detecting AO for large value of ω̂AO. As for IO, the performance of outlier detection procedure is better for model with larger coefficient values. The outlier detection procedure is capable of detecting TC and LC, though the performance is affected if ω is large.",
keywords = "Models, Nonlinear least squares, Outlier detection procedure, Test statistics",
author = "Azami Zaharim and Ibrahim Mohamed and Ibrahim Ahmad and Shahrum Abdullah and Omar, {Mohd. Zaidi}",
year = "2006",
month = "12",
language = "English",
volume = "5",
pages = "1359--1364",
journal = "WSEAS Transactions on Mathematics",
issn = "1109-2769",
publisher = "World Scientific and Engineering Academy and Society",
number = "12",

}

TY - JOUR

T1 - Performances test statistics for single outlier detection in bilinear (1,1,1,1) models

AU - Zaharim, Azami

AU - Mohamed, Ibrahim

AU - Ahmad, Ibrahim

AU - Abdullah, Shahrum

AU - Omar, Mohd. Zaidi

PY - 2006/12

Y1 - 2006/12

N2 - An outlier detection procedure for BL(1,1,1,1) model is developed based on the maxima of the test statistics measuring the effects of IO, AO, TC and LC. A simulation study is carried out in order to investigate the sampling properties of the maxima of the outlier test statistics. It is associated with the sample size, the type of outlier and the coefficients chosen for BL(1,1,1,1). The results show that, in general, the performance of the detection procedure is good. The outlier detection procedure performs well in detecting AO for large value of ω̂AO. As for IO, the performance of outlier detection procedure is better for model with larger coefficient values. The outlier detection procedure is capable of detecting TC and LC, though the performance is affected if ω is large.

AB - An outlier detection procedure for BL(1,1,1,1) model is developed based on the maxima of the test statistics measuring the effects of IO, AO, TC and LC. A simulation study is carried out in order to investigate the sampling properties of the maxima of the outlier test statistics. It is associated with the sample size, the type of outlier and the coefficients chosen for BL(1,1,1,1). The results show that, in general, the performance of the detection procedure is good. The outlier detection procedure performs well in detecting AO for large value of ω̂AO. As for IO, the performance of outlier detection procedure is better for model with larger coefficient values. The outlier detection procedure is capable of detecting TC and LC, though the performance is affected if ω is large.

KW - Models

KW - Nonlinear least squares

KW - Outlier detection procedure

KW - Test statistics

UR - http://www.scopus.com/inward/record.url?scp=33751572472&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33751572472&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33751572472

VL - 5

SP - 1359

EP - 1364

JO - WSEAS Transactions on Mathematics

JF - WSEAS Transactions on Mathematics

SN - 1109-2769

IS - 12

ER -