Rapid profiling of plantation stocks in Bursa Malaysia with Expectation Maximization clustering

Ng, Keng Hoong and Khor, Kok Chin and Tan, Hui Poh (2014) Rapid profiling of plantation stocks in Bursa Malaysia with Expectation Maximization clustering. In: Knowledge Management International Conference (KMICe) 2014, 12 - 15 August 2014, Langkawi, Malaysia.

[img] Text
Rapid Profiling of Plantation Stocks in Bursa Malaysia with Expectation Maximization Clustering.pdf
Restricted to Repository staff only

Download (301kB)
Official URL: http://www.kmice.cms.net.my/kmice2014/AcceptedPape...

Abstract

Building a stock portfolio often requires extensive financial knowledge and Herculean efforts looking at the amount of financial data to analyze. In this preliminary study, the objective is to build a plantation stock portfolio using a clustering technique. We utilized Expectation Maximization (EM) clustering on 38 plantation stocks listed in Bursa Malaysia using 14 financial ratios derived from fundamental analysis. The stocks with the similar financial performance tend to group together and form clusters. This will allow investors to analyze and profile each cluster rapidly. The profiling shall assist the investors in selecting appropriate stocks for their stock portfolio. In this study, the EM algorithm yielded two clusters. The 1-year stock price movement was then used to assess the performance of these two clusters. The result showed that the cluster with a better profile obtained a higher average capital gain as compared with the other cluster.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Plantation stocks, clustering, financial ratios, expectation maximization, expectation-maximization algorithms
Subjects: Q Science > QA Mathematics > QA273-280 Probabilities. Mathematical statistics
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 23 Nov 2016 06:52
Last Modified: 23 Nov 2016 07:29
URI: http://shdl.mmu.edu.my/id/eprint/6377

Actions (login required)

View Item View Item