Citation

Kannan, Rajkumar and Andres, Frederic and Sonai Muthu Anbananthen, Kalaiarasi and Anutariya, Chutiporn (2024) Predicting Gold Prices with Rolling Average Representation and Machine Learning. In: 2024 1st International Conference on Innovative Engineering Sciences and Technological Research (ICIESTR), 14-15 May 2024, Muscat, Oman.

Text Predicting Gold Prices with Rolling Average Representation and Machine Learning.pdf - Published Version Restricted to Repository staff only Download (446kB)

Abstract

Price prediction and forecasting have been important topics that have been long studied in the financial world. Commodity price prediction is one such highly popular application in stock markets across the world. Among all commodities such as gold, silver, platinum and others, gold has continued to be the prime and forefront commodity which decides the price movements of other commodities. Moving average approach and its variants are the important technical indicators which help financial institutions, investors and traders to understand the movement of gold prices throughout the year. Some of the popular technical indicators are simple moving averages, cumulative moving averages and exponential moving averages. This paper proposes a unique gold price prediction methodology by building popular ensemble machine learning models utilizing all three types of moving average values as feature representation and next day price as the target for prediction. The predicted values based on the three moving average features are presented to users, investors and traders, so that investors and traders will know the gold price for the next day. Extensive experiments are conducted on the Indian gold price daily data from December 2011 to March 2024 by comparing six prominent machine learning models and ARIMA model. We also present a detailed study of different feature importance. Experimental results on the gold price daily data have confirmed that polynomial regression models have given superior performance in terms of regression accuracy and error metrics namely MSE, RMSE and MAE

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Machine Learning, Time series data, Gold prices, Moving averages
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	12 Feb 2025 01:53
Last Modified:	12 Feb 2025 01:53
URII:	http://shdl.mmu.edu.my/id/eprint/13428

Downloads

Edit (login required)