Gold Price Prediction Based On Yahoo Finance Data Using Lstm Algorithm

• Main Authors: Dhuhita, Windha Mega Pradnya, Farid, Muhammad Farhan Al, Yaqin, Ainul, Haryoko, Haryoko, Huda, Arif Akbarul
• Format: Conference Proceeding
• Language: English
• Subjects: error; Finance; Fluctuations; Gold; gold price; grid search; Informatics; lstm; Multimedia systems; prediction; Prediction algorithms; Predictive models
• ISSN: 2837-5203
• DOI: 10.1109/ICIMCIS60089.2023.10349035

Gold, a highly valued precious metal, has significant intrinsic value in contemporary society. Therefore, more and more people are embarking on precious metal investments known as gold. Individuals who wish to invest in gold should be vigilant in monitoring the fluctuations in this precious metal's buying and selling prices. Yahoo Finance is a leading online platform that can be used as a reliable reference for monitoring the fluctuations in gold's buying and selling prices. These factors mainly include changes in the closing price, opening price, highest value, and lowest value of gold.The observed phenomenon shows that gold prices exhibit high volatility, mainly due to frequent and repeated fluctuations. The LSTM (Long Short-Term Memory) technique, coupled with applying hyperparameter optimization through grid search, enables accurate gold price prediction using historical gold price data. This result is achieved after the grid search procedure, where the lowest error result is 0.00033, generated by the LSTM parameter unit of 200 with a dropout of 0.1.