Application of artificial neural network in share price forecasting / Zakiah Mustafa

Mustafa, Zakiah (1998) Application of artificial neural network in share price forecasting / Zakiah Mustafa. Degree thesis, Universiti Teknologi MARA (UiTM).

Abstract

Abstract This thesis present the applications of artificial neural network (ANN) to predict the share price of Telekom Malaysia Bhd. The back-propagation algorithm used to train the ANN. The ANN model developed has three layers i.e input layer, hidden layer and output layer. In this projece,seven (7) input variables are chosen as the key factors for the changes of stock price. There are BSKL composite index, BSKL consumer index, RM currency ( based on US dollar), Dow Jones index, Man Seng index, condition of Prime Minister which is categorized with 1 for stable and 0 for not stable and changes of the stock price. The output is the highest price (closing price) for that day. This approach is used for a short-term prediction where actual data is employed in the experiments. The results obtained are compared with the actual data. Back propagation feed-forward connectionist network have been using to do the prediction of Telekom share price for year 1997. The finding from the experiments demonstrate that with more training data and right parameters results in better prediction. This thesis also highlights the importance of determining the optimum parameters of the ANN to obtain good results. From the test carried out, it was found that neural network method can be used to predict the share price of stock market.

Metadata

Item Type: Thesis (Degree)
Creators:
Creators
Email / ID Num.
Mustafa, Zakiah
UNSPECIFIED
Contributors:
Contribution
Name
Email / ID Num.
Advisor
Hamzah, Noraliza
UNSPECIFIED
Divisions: Universiti Teknologi MARA, Shah Alam > Faculty of Electrical Engineering
Programme: Bachelor of Electrical Engineering (Honours)
Date: 1998
URI: https://ir.uitm.edu.my/id/eprint/102988
Edit Item
Edit Item

Download

[thumbnail of 102988.pdf] Text
102988.pdf

Download (459kB)

Digital Copy

Digital (fulltext) is available at:

Physical Copy

Physical status and holdings:
Item Status:

ID Number

102988

Indexing

Statistic

Statistic details