Thing, Cia (2025) Analisis Time-Series untuk Prediksi Harga Emas Menggunakan Metode Deep-Learning: LSTM. Undergraduate thesis, Universitas Pembangunan Jaya.
![[img]](https://eprints.upj.ac.id/style/images/fileicons/text.png) |
Text
PENDAHULUAN.pdf Download (2MB) |
|
Text
ABSTRACT.pdf Download (128kB) |
|
Text
ABSTRAK.pdf Download (142kB) |
|
Text
DAFTAR_ISI.pdf Download (176kB) |
|
Text
DAFTAR_GAMBAR.pdf Download (240kB) |
|
Text
DAFTAR_TABLE.pdf Download (174kB) |
|
Text
DAFTAR_LAMPIRAN.pdf Download (106kB) |
|
Text
BAB_I.pdf Download (457kB) |
|
Text
BAB_II.pdf Download (1MB) |
|
Text
BAB_III.pdf Download (486kB) |
|
Text
BAB_IV.pdf Download (1MB) |
|
Text
BAB_V.pdf Download (5MB) |
|
Text
BAB_VI.pdf Download (188kB) |
|
Text
DAFTAR_PUSTAKA.pdf Download (260kB) |
|
Text
LAMPIRAN.pdf Download (4MB) |
|
Text
BERITA ACARA UNGGAH MANDIRI.pdf Download (283kB) |
|
Text
Bukti Lolos Similarity.pdf Download (11MB) |
Abstract
The price of gold has a complex relationship with financial and macroeconomic factors, making it difficult to forecast. The complexity of price forecasting provides the motivation for developing a model that can predict daily gold prices with high accuracy and the smallest possible error. This study proposes the Long Short-Term Memory (LSTM) method to optimally forecast gold prices. This research uses features such as the opening, closing, highest, and lowest gold prices. External features are also used as additional features, like crude oil prices, silver prices, stock indices, and the United States (US) Dollar Index. Based on the results obtained, LSTM achieved optimal results with the lowest error metrics: an MSE of 165.77, an RMSE of 9.36, an MAE of 12.88, and a MAPE of 0.37%. With these results, the LSTM method proves itself as an effective and reliable methodology for forecasting gold prices. LSTM can be relied upon as a primary choice for forecasting future gold prices, providing an important contribution to the world of finance and investment.
| Item Type: | Karya Tulis Ilmiah (KTI) (Undergraduate) |
|---|---|
| Subjects: | H Social Sciences > HA Statistics H Social Sciences > HG Finance Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QC Physics |
| Divisions: | Fakultas Teknologi dan Desain > Informatika |
| Depositing User: | Cia Thing |
| Date Deposited: | 29 Jul 2025 10:27 |
| Last Modified: | 29 Jul 2025 10:28 |
| URI: | http://eprints.upj.ac.id/id/eprint/11644 |
Actions (login required)
 |
View Item |