Classification of dividend news based on the movement of the share market prices of public listed companies in Bursa Malaysia / Shubana Vijaya Kumar

Shubana , Vijaya Kumar (2020) Classification of dividend news based on the movement of the share market prices of public listed companies in Bursa Malaysia / Shubana Vijaya Kumar. Masters thesis, Universiti Malaya.

[img] PDF (The Candidate's Agreement)
Restricted to Repository staff only

Download (197Kb)
    PDF (Thesis M.A)
    Download (1712Kb) | Preview


      Stock market is naturally complex and plays a major role in towards the nation’s growth. However, the performance of a company in stock market varies due to many influences but not limited to economics, political and financial related news. This study attempts to classify the share market dividend news announcement in Bursa Malaysia based on the pattern of share market price. Samples including five hundred (500) observations of dividend news from forty-seven (47) listed companies in Bursa Malaysia during the period of 2000 to 2018 are used in this study. There are three (3) main objectives in this study which consist of (1) to propose a dividend news classification, (2) to develop a dividend news classification system and (3) to evaluate the accuracy of the news classification approach. The first objective is achieved by proposing a news classification approach to classify the dividend news based on announcement date price, ex-date price, entitlement date price and payment date price. This research uses historic data to analyse and classify the dividend news. The dividend news classification approach is determined by comparing daily closing price difference from the dividend announcement date until dividend payment date and price difference between the four (4) dates - dividend announcement date, ex-date, entitlement date and dividend payment date. The dividend news is classified as a good news if the price difference is more or equal than zero and if it is lower, it is classified as a bad news. Second objective is achieved with the development of the dividend news classification system by imposing the mathematical approach which calculates the price difference between the announcement date, ex-date, entitlement date and payment date. The news classification system is developed by using Joget Workflow which is an open source platform. In addition, the accuracy of the proposed mathematical approach for dividend news classification system is evaluated using five (5) different Machine Learning Model which consist of Logistic Regression (LR), Linear Discriminant Analysis (LDA), K Neighbours Classifiers (KNN), Gaussian Naïve Bayes (NB) and Support Vector Machine (SVM). As per to achieve the third objective of this study, the mathematical approach is strongly supported by Logistic Regression (LR) model with the highest accuracy of 0.98. The outcome of this study helps to facilitate the investors in interpreting the dividend news easily and faster with simplified methods. Finally, it is important to the investors who would like to analyse and study the dividend news trend in the past years.

      Item Type: Thesis (Masters)
      Additional Information: Dissertation (M.A.) – Faculty of Computer Science & Information Technology, Universiti Malaya, 2020.
      Uncontrolled Keywords: Dividend news; News classification; Share market; Machine learning; Bursa Malaysia
      Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
      Q Science > QA Mathematics > QA76 Computer software
      Divisions: Faculty of Computer Science & Information Technology
      Depositing User: Mr Mohd Safri Tahir
      Date Deposited: 10 Feb 2022 06:42
      Last Modified: 10 Feb 2022 06:42

      Actions (For repository staff only : Login required)

      View Item