Enhanced symbolic aggregate approximation algorithm for financial time series data representation

Peiman Mamanibarnaghi (P53642)

Please use this identifier to cite or link to this item: https://ptsldigital.ukm.my/jspui/handle/123456789/476528

Full metadata record

DC Field	Value	Language
dc.contributor.advisor	Azuraliza Abu Bakar, Prof. Dr.	-
dc.contributor.author	Peiman Mamanibarnaghi (P53642)	-
dc.date.accessioned	2023-10-06T09:20:23Z	-
dc.date.available	2023-10-06T09:20:23Z	-
dc.date.issued	2012-06-25	-
dc.identifier.other	ukmvital:115202	-
dc.identifier.uri	https://ptsldigital.ukm.my/jspui/handle/123456789/476528	-
dc.description	Data representation is one of the most important tasks in time series data pre-processing. Time series data representation is required to make the data more suitable for data mining specifically for prediction. Time series data is characterized by its numerical and continuous values. It usually contains large volumes of data collected in seconds, minutes or by the hour. Time series data has been extremely used in having an accurately financial prediction. However, the challenge of the current time series approach is the feet that to reduced dimension of data into appropriate number of interval. Therefore it avoiding sever loss of knowledge by remaining the original min and max values rather than using the mean values as SAX did. One of the most popular data representation methods for time series is the Symbolic Aggregate Approximation (SAX). SAX is based on the approach called Piecewise Aggregate Approximation where data are approximately aggregated in piecewise manner and represented in symbol form. SAX uses mean values as the basis of representation of the data. However, representing the time series financial data with the mean value often causes the loss of patterns that can describe important pieces of information. The aim of this study is to improve the performance of SAX by removing some inconsistency in the original time series data before the predictive model is developed. The study consists of three main phases. The first phase is data collection and preparation. The data contains various exchange rate data at a certain time. The second phase is the implementation of the proposed representation approach called Enhanced-SAX (EnSAX) where the existing SAX is improved. EnSAX adds two new values that are the minimum and maximum value to the original mean value for each segment in SAX. This value enables better representation for each segment in a lower dimension and keeps some of the important patterns that are meaningful in financial time series data. The third phase is predictive modelling using the linear regression and Support Vector Machines (SVM) technique. The experimental results show that the EnSAX representation manages to preserve some important information compared to SAX. EnSAX with some additional values provides a better performance for financial time series prediction and improves predictive accuracy.,“Certification of Master’s/Doctoral Thesis” is not available,Master of Information Technology	-
dc.language.iso	eng	-
dc.publisher	UKM, Bangi	-
dc.relation	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat	-
dc.rights	UKM	-
dc.subject	Time-series analysis	-
dc.subject	Computer algorithms	-
dc.subject	Universiti Kebangsaan Malaysia -- Dissertations	-
dc.subject	Dissertations, Academic -- Malaysia	-
dc.title	Enhanced symbolic aggregate approximation algorithm for financial time series data representation	-
dc.type	theses	-
dc.format.pages	70	-
dc.identifier.callno	QA280.M339 2012 3 tesis	-
dc.identifier.barcode	002547(2012)	-
dc.identifier.barcode	004094(2019)(PL2)	-
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Files in This Item:

File	Description	Size	Format
ukmvital_115202+SOURCE1+SOURCE1.0.PDF Restricted Access		56.2 MB	Adobe PDF	View/Open

Show simple item record Recommend this item