A proposed algorithm of random vector in measuring similarity for network topology of Bursa Malaysia

The stock market is a complex system where the interrelationships between the stocks are complicated because it is in multivariate time series setting which consists of opening, highest, lowest and closing prices. Basically, the Pearson correlation coefficient (PCC) is applied to measure the similarity between two or more univariate time series of stocks. However, the economic information from other variables may inaccurate if the analysis is conducted by applying single variable only. Therefore, multi-dimensional of stocks are considered in this thesis. The similarities between two or more multi-dimensional of stocks are quantified by using Random Vector (RV) coefficient. Based on the preliminary analysis, the computational of RV coefficient is difficult, time-consuming, and tedious when a large number of stocks are involved. Hence, to ease the calculation process and improve the computational efficiency of RV coefficient, an algorithm is proposed. The proposed algorithm is able to measure the similarities among all pairs of stocks in Bursa Malaysia at once. The calculation process of RV coefficient among all pairs of stocks can be shortened and eased as the proposed algorithm consists of time complexity of order of O(n2). The behaviors and interactions among the stocks in Bursa Malaysia are then determined by using the Forest of all possible minimum spanning trees. In this thesis, MK Land Holdings Berhad was found out to be the predominant stock in Bursa Malaysia as it displays a star-like structure and is located at the central hub of the network.

Name dc:type.qualificationname

mphil

Level dc:type.qualificationlevel

masters

Grantor dc:publisher.institution

Universiti Tun Hussein Onn Malaysia

Year dc:date.issued

2018

Author dc:creator

• Lim, San Yee

• QA Mathematics

Language dc:language

en