Stock price index prediction using machine learning

Ian L. Perdana, Rofikoh Rokhim

Research output: Contribution to journalConference articlepeer-review

1 Citation (Scopus)


The decision-making process to choose the stocks to be purchased is one of the challenges faced when buying stock. The method of buying stocks is important because it will determine whether the selected shares will generate profits or losses. This paper will explain how one of machine learning algorithm, long short-term memory will be applied and used to predict the movement of index stock prices day-by-day. Developing a model using the long short-term memory algorithm was chosen because it can classify discrete-time series data such as daily index stock price data. The method that will be used to gather the data is web scraping method on yahoo finance website. The model created will then be evaluated and tested to predict the price of the shares purchased by calculating the evaluation method like Mean Squared Error (MSE), Mean Absolute Error (MAE), Root Means Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). To illustrate the use of the algorithm, a case study using five index stocks data in South East Asian country is conducted. From the results of the research that has been done, it was found that the day with the best results for predicting the JKSE stock price index was the second day, the KLSE stock price index was the second day, the PSEi stock price index was the first day, the SET.BK stock price index was the second day, and STI stock price index is the first day.

Original languageEnglish
Article number020031
JournalAIP Conference Proceedings
Issue number1
Publication statusPublished - 6 Nov 2023
Event4th International Conference on Industrial, Enterprise, and System Engineering: Collaboration of Science, Technology, and Innovation Toward Sustainable Development, ICoIESE 2021 - Bandung, Indonesia
Duration: 16 Dec 2021 → …


Dive into the research topics of 'Stock price index prediction using machine learning'. Together they form a unique fingerprint.

Cite this