<fmt:message key='jsp.layout.header-default.alt'/>  
 

DSpace@UM >
Faculty of Science >
Masters Dissertations : Science >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1812/517

Title: Detecting additive and innovational outliers in BL(p,0,1,1) process
Authors: Mohd Isfahani Ismail
Keywords: BL(p,0,1,1) models
Box-Jenkins
Additive outlier
AO
Innovational outlier
IO
Issue Date: 2009
Publisher: University Malaya
Abstract: This study proposed an outlier detection rocedure for the BL(p,0,1,1) models, where p = 1,2,3. In this process, a time series was first fitted by the models using the Box-Jenkins approach. In the estimation stage, the parameter estimates for the model were found using the nonlinear least squares method. The existence of additive outlier (AO) and innovational outlier (IO) in data from the BL(p,0,1,1) models, p = 1,2,3, were considered in this study. Their features were studied so that the different patterns caused by both type of outliers were distinguishable. Further, the measure of outlier effect for AO and IO were derived using the least square method. Due to the complexity of the statistics, bootstrapping is used to find the variance of the statistics. Based on the bootstrap samples, three different formulae were used to calculate the variance. These formulas are the standard formula, trimmed mean (TM) and MAD. The appropriate test criteria and test statistics to identify the occurrence of outliers were found by standardizing the observed ω giving three different bootstrap-based procedures. These procedures are then compared to the model-based (MB) procedure. The detection of outliers was carried out by examining the maximum value of the standardized statistics of the outlier effects. The outlier detection procedure for identifying the type of outlier at time point t was proposed. Simulation study was carried out to study the performance of the procedure in BL(p,0,1,1) models, p = 1,2,3. It was found out, in general, the proposed procedure performed well in detecting outliers. As for illustration, the proposed procedure was applied on rainfall data and air quality index data.
Description: Dissertation -- Faculty of Science, University of Malaya, 2009.
URI: http://dspace.fsktm.um.edu.my/handle/1812/517
Appears in Collections:Masters Dissertations : Science

Files in This Item:

File Description SizeFormat
mohd isfahani ismail MSc thesis.pdfThesis Text1.33 MBAdobe PDFView/Open


This item is protected by original copyright



Your Tags:

 

  © Copyright 2008 DSpace Faculty of Computer Science and Information Technology, University of Malaya . All Rights Reserved.
DSpace@UM is powered by MIT - Hawlett-Packard. More information and software credits. Feedback