Modelling share prices with ARMA method

Foong, Chui Chian (2012) Modelling share prices with ARMA method. Universiti Malaysia Sabah. (Unpublished)


Download (10MB) | Preview


Investment in stock market such as an essential asset and forecasting is the tool often used by Investor to forecast future share price. Forecasting exhibit the future trend of the share price and underlying the behavior of the trend. Further planning can be made by company according to the trend analysis and maximize profit by modify the behavior now. In this paper, Star Publication (M) Berhad is chosen as the company to do the study and the historical data of the company is used as the data. There is missing value In data collected and estimated by using polynomial order nine. ARMA method is used to do forecasting and the processes contain four phases. This research restricted p+q≤7, hence there is total 35 possible models, and models are selected by using coefficient test. The selected models will test by eight selection criteria to obtain best model for forecasting. Anal phase is goodness of fit which is test on normality and randomness of the residuals. The backward transformation Is applied In order to transform forecast indexes into forecast prices. At the end of the study, the MAPE is applied to test the reliable of forecasting result. The main finding in this research paper is ARMA method is performed well in short term. In this research demonstrated that residuals is not normality distributed which indicate there is possible source of error that the ADF test does not always produce precise result.

Item Type: Academic Exercise
Uncontrolled Keywords: forecasting, polynomial order nine, ARMA method, share price, investment
Subjects: Q Science > QA Mathematics
Divisions: SCHOOL > School of Science and Technology
Depositing User: ADMIN ADMIN
Date Deposited: 03 Aug 2016 06:15
Last Modified: 20 Nov 2017 03:42

Actions (login required)

View Item View Item