Dimensional ranking and reduction approach for finding optimal stock price influencing factors: An empirical study

• Main Authors: Ahmad, Faudziah, Ku-Mahamud, Ku Ruhana, Mohamed Din, Aniza, Mohamad Mohsin, Mohamad Farhan
• Format: Conference or Workshop Item
• Language: English
• Published: 2011
• Subjects: QA76 Computer software
• Online Access: http://repo.uum.edu.my/10740/1/063.pdf; http://repo.uum.edu.my/10740/; http://ieeexplore.ieee.org/Xplore/home.jsp
• Institution: Universiti Utara Malaysia

A limited number of factors in critical areas are necessary to identify the sustainability of companies [30].It was found that three factors were commonly used [2] and a maximum of seven factors was recommended [2].In order to identify the optimal factors, we propose a method based on rough set theory that consisted of five main steps.These were data cleansing and preparation, dimensional reduction, ranking selected factors, optimal factor extraction, and rules generation and evaluation. Dimensional reduction was conducted using rough set theory where the set of factors were identified from the set of reducts that produce highest classification accuracy.These factors were first ranked through computation of its occurrences in reduct sets and then selected based on highest occurrences.The major contribution of this work is the reduced factors shows competitive results in classifying new cases and therefore keeping the quality of knowledge.