skip to main content

PREDIKSI HARGA EMAS DUNIA MENGGUNAKAN METODE LONG-SHORT TERM MEMORY

*Tania Giovani Lasijan  -  Departemen Statistika, Fakultas Sains dan Matematika, Undip, Indonesia
Rukun Santoso  -  Departemen Statistika, Fakultas Sains dan Matematika, Undip, Indonesia
Arief Rachman Hakim  -  Departemen Statistika, Fakultas Sains dan Matematika, Undip, Indonesia
Open Access Copyright 2023 Jurnal Gaussian under http://creativecommons.org/licenses/by-nc-sa/4.0.

Citation Format:
Abstract
Gold investment is one of the investments that is quite lot of interest by the public and also is considered safer because it has relatively low risk and tends to be stable compared to other investment instruments, especially amid the uncertainty of global economic conditions caused by the COVID-19 pandemic. Awareness about gold price predictions can provide information to people who want to invest in gold so they have higher opportunity to earn profits and minimize the risks obtained. The gold prices prediction method used in this study is Long-Short Term Memory (LSTM) using RStudio. LSTM is one of the method that is widely used to predict time series data. LSTM is a variation of the Recurrent Neural Network (RNN) that is used as a solution to overcome the occurrence of exploding gradient or vanishing gradient in RNN when processing long sequential data. The best LSTM model in this study for predicting gold prices is  the model with MAPE value 2,70601, which is a model with a training data and testing data comparison 70% : 30% and hyperparameters batch size 1, units 1, AdaGrad optimizer, and learning rate 0,1 with 500 epochs.

Note: This article has supplementary file(s).

Fulltext View|Download |  Research Instrument
Untitled
Subject
Type Research Instrument
  Download (160KB)    Indexing metadata
Keywords: Gold; Long-Short Term Memory; Recurrent Neural Network

Article Metrics:

  1. BPS. (2020). Pertumbuhan Ekonomi Indonesia Triwulan III-2020. Berita Resmi Statistik, No. 15/02/(15), 1–12
  2. Chung, H., & Shin, K. S. (2018). Genetic algorithm-optimized long short-term memory network for stock market prediction. Sustainability (Switzerland), 10(10), 1–18. https://doi.org/10.3390/su10103765
  3. Goodfellow, I., Bengio, Y., & Courville, A. (2014). Deep Learning. In MIT Press. MIT Press. http://files.sig2d.org/sig2d14.pdf#page=5
  4. Hochreiter, S., & Schmidhuber, J. (1997b). Long Short-Term Memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.173
  5. Ioffe, S., & Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. International Conference on Machine Learning, 448–456
  6. Julpan, Nababan, E. B., & Zarlis, M. (2015). Analisis Fungsi Aktivasi Sigmoid Biner Dan Sigmoid Bipolar Dalam Algoritma Backpropagation Pada Prediksi Kemampuan Siswa. Jurnal Teknovasi, 02(1), 103–116
  7. Makridakis, S., Wheelwright, S. C., & McGee, V. E. (1999). Metode dan aplikasi peramalan. Jakarta: Erlangga
  8. Manaswi, N. K., & John, S. (2018). Deep Learning with Applications Using Python. Springer
  9. Mary, P. M., Pushpa, R., & Manimala, K. (2014). Implementation of Hyperbolic Tangent Activation Function in VLSI. 2(March), 225–22
  10. Napompech, K., Tanpipat, A., & Nidpa, U. (2010). Factors Influencing Gold Consumption for Savings and Investments by People in the Bangkok Metropolitan Area. International Journal of Arts and Sciences, 3(7), 508–520. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.301.2007&rep=rep1&type=pdf
  11. Zhao, Z., Chen, W., Wu, X., Chen, P. C. Y., & Liu, J. (2017). LSTM network: a deep learning approach for short-term traffic forecast. IET Intelligent Transport Systems, 11(2), 68–75

Last update:

No citation recorded.

Last update:

No citation recorded.