PREDIKSI HARGA EMAS DENGAN METODE LONG SHORT-TERM MEMORY

No Thumbnail Available

Date

2023-11

Journal Title

Journal ISSN

Volume Title

Publisher

Elfitra

Abstract

Predicting gold prices holds significant importance in the realms of finance and investment, enabling market participants to make wiser decisions amidst gold price fluctuations. The training results of the Long Short-Term Memory (LSTM) model displayed remarkable performance, with a Mean Squared Error (MSE) value of 0.0034 and a Mean Absolute Percentage Error (MAPE) value of 7.13%. According to the MAPE criterion, this LSTM model's predictive capabilities can be categorized as highly accurate. These outcomes affirm the potential of the LSTM model in providing precise gold price predictions, aiding market participants in more informed decision-making. The forecasted gold prices using LSTM from January 1, 2023, to March 1, 2023, indicate a daily decrease in gold prices. Consequently, a prudent approach for market participants would be to engage in future gold purchases during the predicted gold price decrease in 2023.

Description

Keywords

Data Scaling, Gold Price Prediction, LSTM, MAPE, MSE

Citation

Perpustakaan

Collections