Comparative Analysis of ARIMA and Transformer-Based Models for Gold Price Forecasting
DOI:
https://doi.org/10.54097/6cjkdf11Keywords:
ARIMA, Transformer, Gold Price Forecasting, Financial Time Series, Deep Learning.Abstract
In recent years, the powerful capabilities of deep learning models have attracted considerable attention from both financial investors and researchers. Emerging architectures such as Transformer and its variants (e.g., Autoformer, Informer) have demonstrated superior ability in capturing nonlinear patterns within long-term time series, leading to their widespread application in tasks such as forecasting financial market indices. However, it remains unclear whether such models are universally effective across all financial prediction scenarios. In this study, we focus on a specific domain of financial forecasting—gold price prediction—and conduct extensive experiments on two datasets. The results indicate that traditional statistical models, such as ARIMA, achieve higher predictive accuracy than deep learning models. This finding highlights the importance of carefully considering the inherent characteristics of time-series data. In contexts such as gold price prediction, where short-term linear patterns dominate, simple statistical model may prove more effective.
Downloads
References
[1] Qian Y, Ralescu D A, Zhang B. The analysis of factors affecting global gold price[J]. Resources Policy, 2019, 64: 101478.
[2] Toraman C, Basarir C, Bayramoglu M F. Determination of factors affecting the price of gold: A study of MGARCH model[J]. Business and economics research journal, 2011, 2(4): 37-50.
[3] Shumway R H, Stoffer D S. ARIMA models[M]//Time series analysis and its applications: with R examples. Cham: Springer International Publishing, 2017: 75-163.
[4] Han K, Wang Y, Chen H, et al. A survey on vision transformer[J]. IEEE transactions on pattern analysis and machine intelligence, 2022, 45(1): 87-110.
[5] Zhou H, Zhang S, Peng J, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//Proceedings of the AAAI conference on artificial intelligence. 2021, 35(12): 11106-11115.
[6] Wu H, Xu J, Wang J, et al. Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting[J]. Advances in neural information processing systems, 2021, 34: 22419-22430.
[7] Kitaev N, Kaiser Ł, Levskaya A. Reformer: The efficient transformer[J]. arXiv preprint arXiv:2001.04451, 2020.
[8] Sims C A. Macroeconomics and reality[J]. Econometrica, 1980, 48(1): 1–48.
[9] Bollerslev T. Generalized autoregressive conditional heteroskedasticity[J]. Journal of econometrics, 1986, 31(3): 307-327.
[10] Box G E P, Jenkins G M, Reinsel G C, Ljung G M. Time series analysis: forecasting and control[M]. 5th ed. John Wiley & Sons, 2015. 11
[11] Bollerslev T. Generalized autoregressive conditional heteroskedasticity[J]. Journal of Econometrics, 1986, 31(3): 307–327.
[12] Mandelbrot B. The variation of certain speculative prices[J]. The Journal of Business, 1963, 36(4): 394–419.
[13] Cont R. Empirical properties of asset returns: stylized facts and statistical issues[J]. Quantitative Finance, 2001, 1(2): 223–236.
[14] Taleb N N. The Black Swan: The Impact of the Highly Improbable[M]. Random House, 2007.
[15] Kristjanpoller W, Minutolo M C. Gold price volatility: A forecasting approach using the Artificial Neural Network–GARCH model[J]. Expert Systems with Applications, 2015, 42(20): 7245–7251.
[16] Arouri M E H, Lahiani A, Nguyen D K. Forecasting the conditional volatility of oil spot and futures prices with structural breaks and long memory models[J]. Energy Economics, 2012, 34(1): 283–293.
[17] Rumelhart D E, Hinton G E, Williams R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088): 533–536.
[18] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735–1780.
[19] Bengio Y, Simard P, Frasconi P. Learning long-term dependencies with gradient descent is difficult[J]. IEEE Transactions on Neural Networks, 1994, 5(2): 157–166.
[20] Siami-Namini S, Tavakoli N, Namin A S. A comparative analysis of forecasting financial time series using ARIMA, LSTM, and BiLSTM[EB/OL]. arXiv:1811.11940, 2018.
[21] Sutskever I, Vinyals O, Le Q V. Sequence to sequence learning with neural networks[C]//Advances in Neural Information Processing Systems. 2014.
[22] Bahdanau D, Cho K, Bengio Y. Neural machine translation by jointly learning to align and translate[EB/OL]. arXiv:1409.0473, 2014.
[23] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems. 2017.
[24] Fischer T, Krauss C. Deep learning with long short-term memory networks for financial market predictions[J]. European Journal of Operational Research, 2018, 270(2): 654–669.
[25] Bao W, Yue J, Rao Y. A deep learning framework for financial time series using stacked autoencoders and long-short term memory[J]. PLoS One, 2017, 12(7): e0180944.
[26] Livieris I E, Pintelas E, Pintelas P. A CNN–LSTM model for gold price time-series forecasting[J]. Neural Computing and Applications, 2020, 32(23): 17351–17360.
[27] Devlin J, Chang M W, Lee K, Toutanova K. BERT: Pre-training of deep bidirectional transformers for language understanding[EB/OL]. arXiv:1810.04805, 2018.
[28] Li S, Jin X, Xuan Y, Zhou X, Chen W, Wang Y X, Yan X. Enhancing the locality and breaking the memory bottleneck of transformer on time series forecasting[C]//Advances in Neural Information Processing Systems. 2019.
[29] Zhou H, Zhang S, Peng J, Zhang S, Li J, Xiong H, Zhang W. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(12): 11106–11115.
[30] Lim B, Arık S Ö, Loeff N, Pfister T. Temporal fusion transformers for interpretable multi-horizon time series forecasting[J]. International Journal of Forecasting, 2021, 37(4): 1748–1764.
[31] Ding Q, Wu S, Sun H, Guo J, Guo J. Hierarchical transformer with contrastive learning for long-term time series forecasting[J]. Expert Systems with Applications, 2023, 215: 119343.
[32] Xiong R, Yang Y, He D, Zheng K, Zheng S, Xing C, et al. On layer normalization in the transformer architecture[C]//International Conference on Machine Learning. 2023.
[33] Hamilton J D. Time series analysis[M]. Princeton university press, 2020.
[34] Chatfield C, Xing H. The analysis of time series: an introduction with R[M]. Chapman and hall/CRC, 2019.
[35] Box G E P, Jenkins G M, Reinsel G C, et al. Time series analysis: forecasting and control[M]. John Wiley & Sons, 2015.
[36] Hirschberg J, Manning C D. Advances in natural language processing[J]. Science, 2015, 349(6245): 261-266.
[37] Elman J L. Finding structure in time[J]. Cognitive science, 1990, 14(2): 179-211.
[38] LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 2002, 86(11): 2278-2324.
[39] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30.
[40] Alammar J. The illustrated transformer[J]. The Illustrated Transformer–Jay Alammar–Visualizing Machine Learning One Concept at a Time, 2018, 27(1).
[41] Zhou H, Zhang S, Peng J, et al. Informer: Beyond efficient transformer for long sequence time-series forecasting[C]//Proceedings of the AAAI conference on artificial intelligence. 2021, 35(12): 11106-11115.
[42] Chatfield C, Xing H. The analysis of time series: an introduction with R[M]. Chapman and hall/CRC, 2019.
[43] Indyk P, Motwani R. Approximate nearest neighbors: towards removing the curse of dimensionality[C]//Proceedings of the thirtieth annual ACM symposium on Theory of computing. 1998: 604-613.
[44] Gomez J, Barnett M A, Natu V, et al. Microstructural proliferation in human cortex is coupled with the development of face processing[J]. Science, 2017, 355(6320): 68-71.
[45] beekiran00. Gold-Price-prediction-Excel-and-R[CP/OL]. GitHub, 2022-01-23 [2025-09-11]. https://github.com/beekiran00/Gold-Price-prediction-Excel-and-R.
[46] Yanke Huang X Z. Analyzing Trends in Trading Patterns in Financial Markets Using Deep Learning Algorithms [J]. Journal of Electrical Systems, 2024, 20(3s): 1542–1555.
[47] Sahoo M, Nayak P P, Hanhaga M, et al. Exploring the asymmetric effect of remittance inflows on gold import demand: Evidence from a large gold-consuming and remittance-receiving Country [J]. Resources Policy, 2023, 85: 103900.
[48] M Dr K K, Palaniappan N, Sultana R, et al. The Impact of Macroeconomic Factors on Gold Prices [J]. INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT, 2024, 08(12): 1–5.
[49] Cheng X. A Comprehensive Study of Feature Selection Techniques in Machine Learning Models [J]. Insights in Computer, Signals and Systems, 2024, 1(1): 65–78.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Academic Journal of Management and Social Sciences
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
