DEVELOPMENT OF STOCK MARKET PREDICTION MOBILE SYSTEM IN BLUE CHIP STOCKS FOR MALAYSIA SHARE MARKET USING DEEP LEARNING TECHNIQUE

• Main Authors: Chong, Fong Kim, Yong, Sik Tian, Yap, Choi Sen
• Format: Article
• Language: English
• Published: INTI International University 2020
• Subjects: L Education (General); QA75 Electronic computers. Computer science
• Online Access: http://eprints.intimal.edu.my/1460/1/ij2020_42.pdf; http://eprints.intimal.edu.my/1460/; http://intijournal.newinti.edu.my

Bursa Malaysia is the stock market of Malaysia where the exchange is tracked by the Kuala Lumpur Composite Index (KLCI) and blue chip stocks are the stocks trading in KLCI as well. Blue chip stocks are stocks issued by well-established and market capitalization firms, which have a sound financial performance for an extended period. There are various techniques investors use in the stock market investment; some may use fundamental analysis, technical analysis, emotion influence or even gambling technique. None of the mentioned techniques guarantee of 100% profit in stock market, which because of low accuracy analysis, lack of knowledge with no proper study on the stock, casino mentality in the stock market or even with no proper investment goal. Most of the Malaysian is not interested to invest in the stock market due to risk of losing money. This paper will look into the use of deep learning technique in developing a stock prediction system in mobile android platform with the features of predicting and recommending stock price mainly for blue chip stock in Malaysia Stock Market. Therefore, the objective of this paper is to look into the used of Long-Short Term Memory (LSTM), one of the deep learning technique applied in the prototype system, which to improve the accuracy of forecasting in stock market in term of stock price prediction and the recommendation of the buy or sell mode for the 30 samples blue chip stocks. According to Isah and Zulkermine (2019), the accuracy of stock prediction is about 72% to 85% and the prediction can be made successfully with LSTM (Seyda, Akhtar, etc. 2020).