Improvement on Additive Outlier Detection Procedure in Bilinear Model
Article Sidebar
Main Article Content
Abstract
This paper considers the problem of outlier detection in bilinear time series data; with
special focus on two most basic models BL(1,0,1,1) and BL(1,1,1,1). The formulation of effect of additive outlier on the observations and residuals has been developed and the least squares estimator of the outlier effect has been derived. Consequently, an outlier detection procedure employing bootstrapping method to estimate the variance of the estimator has been proposed. In this paper, we propose to use the mean absolute deviance and trimmed mean methods to improve the performances of the procedure. Using simulation works, we show that trimmed method has successfully improved the performance. Subsequently the procedure is applied to a real data set
This paper considers the problem of outlier detection in bilinear time series data; with
special focus on two most basic models BL(1,0,1,1) and BL(1,1,1,1). The formulation of effect of additive outlier on the observations and residuals has been developed and the least squares estimator of the outlier effect has been derived. Consequently, an outlier detection procedure employing bootstrapping method to estimate the variance of the estimator has been proposed. In this paper, we propose to use the mean absolute deviance and trimmed mean methods to improve the performances of the procedure. Using simulation works, we show that trimmed method has successfully improved the performance. Subsequently the procedure is applied to a real data set
This paper considers the problem of outlier detection in bilinear time series data; with
special focus on two most basic models BL(1,0,1,1) and BL(1,1,1,1). The formulation of effect of additive outlier on the observations and residuals has been developed and the least squares estimator of the outlier effect has been derived. Consequently, an outlier detection procedure employing bootstrapping method to estimate the variance of the estimator has been proposed. In this paper, we propose to use the mean absolute deviance and trimmed mean methods to improve the performances of the procedure. Using simulation works, we show that trimmed method has successfully improved the performance. Subsequently the procedure is applied to a real data set
Downloads
Article Details
Licensee MJS, Universiti Malaya, Malaysia. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).