多源异构数据图像整合预测方法研究——以黄金价格预测为例

doi:10.12341/jssms22104

黄金是一种拥有商品、货币、投资等多重属性的特殊大宗商品,而且黄金市场的交易范围和参与者涉及全球,所以其价格会受到多种因素的影响.同时,由于这些影响因素来源分散,数据异构性强并且有各自的规律可循,所以黄金价格本身具有复杂性、非平稳性以及非线性等特征,这也直接导致了预测黄金价格的变动趋势非常具有挑战性.基于黄金的这一复杂特性,文章提出一种全新的基于图像挖掘技术的多源数据融合预测方法— MODII (Multiple Original Data Images Integration).通过2种数据可视化方法将黄金价格的不同影响要素数据转化为4种类型的图像,将这些来自不同市场的不同影响因素通过图像整合在一起,并利用卷积神经网络(Convolutional Neural Network,CNN)的多通道输入结构将多种类型的图像进行融合,之后由CNN自动学习各图像特征及所占权重并进行预测,样本外预测准确率可达81.06%,一定程度上提高了预测准确率.

关键词： 卷积神经网络, 图像挖掘, 趋势预测, 多源数据, 黄金价格

Gold is a special commodity with multiple attributes, such as commodity, currency and investment. With wide range of trade and participants across the globe, gold price may be affected by various factors. Furthermore, scattered sources of these influencing factors, coupled with strong heterogeneity of the data structure with respective regularity makes gold price complex, non-stationary and non-linear, which directly leads to the difficulties of predicting the trend of the gold price. By leveraging the complex characteristics of the gold price, this paper proposed a new multi-source data fusion prediction method based on image mining technology — MODII (Multiple Original Data Images Integration) to accurately predict gold price. Specifically, we collected various of influencing factors of the gold price across markets and transformed them into four types of images. Then, we built a convolutional neural network (CNN) by taking these cross-batch images as input to learn the features and predict corresponding gold price. Our method could reach 81.06% accuracy for out-of-sample prediction and outperform the state-of-the-art methods.

Key words: Convolutional neural network, image mining, trend prediction, multi-source data, gold price

MR(2010)主题分类:

91B25
97R40

分享此文:

[1] Canziani A, Paszke A, Culurciello E. An analysis of deep neural network models for practical applications. 29th Conference on Neural Information Processing Systems (NIPS 2016), ArXiv: 1605.07678, 2017.
[2] Chen C H. Neural networks for financial market prediction. Proceedings of 1994 IEEE International Conference on Neural Networks. Orlando, FL, USA: IEEE, 1994, 1199-1202.
[3] Zhang G P. Time series forecasting using a hybrid ARIMA and neural network model. Neurocomputing, 2003, 50: 159-175.
[4] 张焱, 欧阳一鸣, 王浩, 等. 数据挖掘在金融领域中的应用研究. 计算机工程与应用, 2004, (18): 208-211. (Zhang Y, Ouyang Y M, Wang H, et al. Application of data mining in the financial field. Computer Engineering and Applications, 2014, (18): 208-211.)
[5] Parisi A, Parisi F, Diaz D. Forecasting gold price changes: Rolling and recursive neural network models. Journal of Multinational Financial Management, 2008, 18(5): 477-487.
[6] Kristjanpoller W, Minutolo M C. Gold price volatility: A forecasting approach using the artificial neural network-GARCH model. Expert Systems with Applications, 2015, 42(20): 7245-7251.
[7] 李梦泽, 朱冬元. 基于深度学习集成方法(SDAE-B)模型的黄金价格走势预测实证分析. 中国资产评估, 2021, (1): 12-19. (Li M Z, Zhu D Y. Empirical analysis of gold price prediction based on deep learning integrated method (SDAE-B) model. Appraisal Journal of China, 2021, (1): 12-19.)
[8] 韩旭, 杨珊, 王喜梅. 基于商空间的黄金价格SVM模型预测. 黄金科学技术, 2020, 1(28): 148-157. (Han X, Yang S, Wang X M. SVM model prediction of gold price based on quotient space. Gold Science and Technology, 2020, 1(28): 148-157.)
[9] 郑鹏. 基于改进的PEEMD-BPNN模型的黄金价格走势研究. 学位论文. 成都理工大学, 成都, 2019. (Zheng P. Research on gold price trend based on improved PEEMD-BPNN model. Master Thesis. Chengdu University of Technology, Chengdu, 2019.)
[10] Li Y, Wang S, Wei Y, et al. A new hybrid VMD-ICSS-BiGRU approach for gold futures price forecasting and algorithmic trading. IEEE Transactions on Computational Social Systems, 2021, 1-12.
[11] Sezer O B, Gudelek M U, Ozbayoglu A M. Financial time series forecasting with deep learning: A systematic literature review: 2005-2019. Applied Soft Computing, 2020, 90: 106181.
[12] 刘璐, 娄磊, 刘先俊, 等. 基于LSTM的COMEX黄金期货价格预测对比研究. 长春理工大学学报(自然科学版), 2021, 44(2): 129-135. (Liu L, Lou L, Liu X J, et al. A comparative study of COMEX gold futures price forecast based on LSTM. Journal of Changchun University of Science and Technology $($Natural Science Edition), 2021, 44(2): 129-135.)
[13] Troiano L, Villa E M, Loia V. Replicating a trading strategy by means of LSTM for financial industry applications. IEEE Transactions on Industrial Informatics, 2018, 14(7): 3226-3234.
[14] 孙舒蓉. 基于k线图机器识别的趋势预测及量化策略设计. 硕士论文. 上海师范大学, 上海, 2020. (Sun S R. Trend prediction and quantitative strategy design based on k-graph machine recognition. Master Thesis. Shanghai Normal University, Shanghai, 2020.)
[15] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems $($NIPS$)$, Communications of the ACM, 2017, 60(6): 84-90.
[16] Karpathy A, Toderici G, Shetty S, et al. Large-scale video classification with convolutional neural networks. IEEE Conference Publication, 2014.
[17] Lawrence S, Giles C L, Tsoi A C, et al. Face recognition: A convolutional neural-network approach. IEEE Transactions on Neural Networks, 1997, 8(1): 98-113.
[18] Ciresan D C, Meier U, Gambardella L M, et al. Convolutional neural network committees for handwritten character classification. International Conference on Document Analysis and Recognition, 2011.
[19] Kim Y. Convolutional neural networks for sentence classification. Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar: Association for Computational Linguistics, 2014, 1746-1751.
[20] Kalchbrenner N, Grefenstette E, Blunsom P. A convolutional neural network for modelling sentences. Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, Maryland: Association for Computational Linguistics, 2014, 655-665.
[21] Doering J, Fairbank M, Markose S. Convolutional neural networks applied to high-frequency market microstructure forecasting. 2017 9th Computer Science and Electronic Engineering (CEEC), Colchester: IEEE, 2017, 31-36.
[22] Gunduz H, Yaslan Y, Cataltepe Z. Intraday prediction of Borsa istanbul using convolutional neural networks and feature correlations. Knowledge-Based Systems, 2017, 137: 138-148.
[23] Sezer O B, Ozbayoglu A M. Algorithmic financial trading with deep convolutional neural networks: Time series to image conversion approach. Applied Soft Computing, 2018, 70: 525-538.
[24] Dingli A, Fournier K S. Financial time series forecasting — A deep learning approach. International Journal of Machine Learning and Computing, 2017, 7(5): 118-122.
[25] Vidal A, Kristjanpoller W. Gold volatility prediction using a CNN-LSTM approach. Expert Systems with Applications, 2020, 157: 113481.
[26] 惠国保. 一种基于深度学习的多源异构数据融合方法. 现代导航, 2017, 3(8): 218-223. (Hui G B. Method of multi-source heterogeneous data fusion based on deep learning. Modern Navigation, 2017, 3(8): 218-223.)
[27] 耿立校, 刘丽莎, 李恒昱. 多源异构数据融合驱动的股票指数预测研究. 计算机工程与应用, 2021, 57(20): 142-149. (Geng L X, Liu L S, Li H Y. Research on stock index prediction driven by multi-source heterogeneous data fusion. Computer Engineering and Applications, 2021, 57(20): 142-149.)
[28] 姜建华, 洪年松, 张广云. 一种多源异构数据融合方法及其应用研究. 电子设计工程, 2016, 24(12): 33-36. (Jiang J H, Hong N S, Zhang G Y. Research on multi-source heterogeneous data fusion and its application. Electronic Design Engineering, 2016, 24(12): 33-36.)
[29] 刘运. 基于循环神经网络的多源异构大数据融合模型构建.内蒙古民族大学学报(自然科学版), 2021, 36(3): 204-210. (Liu Y. Construction of a multi-source heterogeneous big data fusion model based on cyclic neural network. Journal of Inner Mongolia University for Nationalities, 2021, 36(3): 204-210.)
[30] Jiang J, Kellly B T, Xiu D. (Re-)Imag(in)ing Price Trends. Chicago Booth Research Paper No. 21-01, Available at SSRN: https://ssrn.com/abstract=3756587 or http://dx.doi.org/10.2139/ssrn.3756587, 2020.
[31] Zhao L T, Wang Y, Guo S Q, et al. A novel method based on numerical fitting for oil price trend forecasting. Applied Energy, 2018, 220: 154-163.
[32] Chen Y C, Rogoff K S, Rossi B. Can exchange rates forecast commodity prices?. The Quarterly Journal of Economics, 2010, 125: 1145-1194.
[33] Reboredo J C. Is gold a safe haven or a hedge for the US dollar?. Journal of Banking & Finance, 2013, 37: 2665-2676.
[34] Mo B, Nie H, Jiang Y. Dynamic linkages among the gold market, US dollar and crude oil market. Physica A: Statistical Mechanics and Its Applications, 2018, 491: 984-994.
[35] Batten J A, Ciner C, Lucey B M. The macroeconomic determinants of volatility in precious metals markets. Resources Policy, 2010, 35: 65-71.
[36] Shafiee S, Topal E. An overview of global gold market and gold price forecasting. Resources Policy, 2010, 35(3): 178-189.
[37] Behmiri N B, Manera M. The role of outliers and oil price shocks on volatility of metal prices. Resources Policy, 2015, 46: 139-150.
[38] Ewing B T, Malik F. Volatility transmission between gold and oil futures under structural breaks. International Review of Economics & Finance, 2013, 25: 113-121.
[39] Mensi W, Beljid M, Boubaker A, et al. Correlations and volatility spillovers across commodity and stock markets: Linking energies, food, and gold. Economic Modelling, 2013, 32: 15-22.
[40] Malliaris A G, Malliaris M. What drives gold returns? A decision tree analysis. Finance Research Letters, 2015, 13: 45-53.
[41] Yurdakul F, Sefa M. An econometric analysis of gold prices in turkey. Procedia Economics and Finance, 2015, 23: 77-85.
[42] Tully E, Lucey B M. A power GARCH examination of the gold market. Research in International Business and Finance, 2007, 21(2): 316-325.
[43] Han A, Hong Y, Wang S. Autoregressive conditional models for interval-valued time series data. The 3rd International Conference on Singular Spectrum Analysis and Its Applications, 2012.
[44] Haque M A, Topal E, et al. Relationship between the gold price and the Australian Dollar-US Dollar exchange rate. Mineral Economics, 2015, 28: 65-78.
[45] Ho R. Handbook of Univariate and Multivariate Data Analysis and Interpretation with SPSS. Boca Raton: Chapman & Hall/CRC, 2006.
[46] Molugaram K, Rao G S. Chapter 6-Correlation and regression. Statistical Techniques for Transportation Engineering, 2017, 293-329.
[47] Zhang P, Ci B. Deep belief network for gold price forecasting. Resources Policy, 2020, 69: 101806.
[48] Wang X, Ma Y, Li W. The prediction of gold futures prices at the Shanghai futures exchange based on the MEEMD-CS-Elman model. SAGE Open, 2021, 11(1): 215824402110018.
[49] Al-Dhuraibi W A, Ali J. Using classification techniques to predict gold price movement. 2018 4th International Conference on Computer and Technology Applications (ICCTA), Istanbul: IEEE, 2018, 127-130.

[1]	郭晓春, 马玉梅, 曹萍萍. 网络环境下考虑目标顾客偏好的服务产品选择方法[J]. 系统科学与数学, 2022, 42(7): 1769-1787.
[2]	王鑫, 王莹. 基于LSTM-CNN的中小企业信用风险预测[J]. 系统科学与数学, 2022, 42(10): 2698-2711.
[3]	王秋雨, 张举勇. 一种基于三维对齐方式的深度学习人脸识别算法[J]. 系统科学与数学, 2021, 41(7): 2035-2045.
[4]	王溪，明扬，洪奕光. 基于群论的深度卷积网络分析[J]. 系统科学与数学, 2019, 39(2): 219-227.

阅读次数

全文

摘要

多源异构数据图像整合预测方法研究——以黄金价格预测为例

Convolutional Neural Network Applied to Gold Price Forecasting with an Image Integration Methods Based on Multi-Sources and Heterogeneous Data