به‏ کارگیری مدل‌ داده‌کاوی هیبریدی (الگوریتم ژنتیک-موجک- شبکة عصبی عمیق- شبیه‌سازی مونت‌کارلو) برای پیش‌بینی قیمت محصولات کشاورزی: مطالعة موردی قیمت آتی زعفران در بورس کالای کشاورزی

نوع مقاله : مقاله پژوهشی

نویسندگان

1 نویسندة مسئول و استادیار پژوهشی، مؤسسة پژوهش‏های برنامه‌ریزی، اقتصاد کشاورزی و توسعة روستایی، تهران، ایران

2 پژوهشگر، مؤسسة پژوهش‏های برنامه‌ریزی، اقتصاد کشاورزی و توسعة روستایی، تهران، ایران.

چکیده

پیش ­بینی قیمت و روند تغییرات آن از مهم­ترین عوامل در تصمیم­ گیری و تدوین راهبرد‌های مربوط به محصولات کشاورزی است. هدف مطالعة حاضر ارائه یک مدل یا الگوی داده‌کاوی هیبریدی شامل مجموعه مدل‌های غیرخطی الگوریتم ژنتیک، تبدیل موجک، شبکه عصبی عمیق و روش مونت‌کارلو برای پیش‌بینی دقیق قیمت محصولات کشاورزی بود. این الگوی پیشنهادی از نوع هیبریدی دومرحله‌ای و مدل پایه هیبریدی غیرخطی- غیرخطی بود و در آن، از الگوریتم ژنتیک برای تعیین وقفه بهینه سری زمانی قیمت، از تابع موجک برای نوفه‌زدایی داده‌های قیمت، از شبکه عصبی عمیق برای پیش‌بینی قیمت، از روش مونت‌کارلو برای شبیه‌سازی محتمل‌ترین احتمال قیمت و در نهایت، از محاسبات پیچیده نرم برای انجام «پیش‌بینی خارج از نمونه با مجموعه داده‌های جدید» برای دوره زمانی دوم تا دهم اردیبهشت 1399 استفاده شد. نتایج مقایسه الگوی پیشنهادی «الگوریتم ژنتیک- تبدیل موجک- شبکه عصبی عمیق- مونت‌کارلو» با سه الگوی رقیب «الگوریتم ژنتیک-شبکه عصبی عمیق- مونت‌کارلو»، «الگوریتم ژنتیک- تبدیل موجک- شبکه عصبی ساده- مونت‌کارلو» و «الگوریتم ژنتیک- شبکه عصبی ساده- مونت کارلو»، با استفاده از معیارهای ارزیابی، نشان داد که الگوی پیشنهادی نسبت به سه الگوی رقیب دارای عملکرد بهتری در پیش‌بینی قیمت زعفران آتی است؛ همچنین، استفاده از شبکه عصبی عمیق در مقایسه با شبکه عصبی ساده و نیز به‏ کارگیری نظریة موجک برای نوفه‌زدایی و استفاده از روش مونت‏کارلو برای شبیه‌سازی قیمت‌های پیش‌بینی‏ شده دقت پیش‌بینی قیمت آتی زعفران را افزایش می‌دهد. علاوه بر این، استفاده از محاسبات نرم برای انجام «پیش‌بینی خارج از نمونه با مجموعه داده‌های جدید» نشان داد که الگوی پیشنهادی از کارآیی لازم و دقت بالا برای پیش‌بینی کوتاه‌مدت قیمت آتی زعفران برخوردار بوده، به‏ گونه ‏ای که میزان خطای محاسباتی کمتر از یک درصد (6/0 درصد) است. بنابراین، مطالعه حاضر در دستیابی به شاخص میزان دقت حداکثری، سناریوسازی روند قیمت‌های آتی، تحلیل حساسیت مؤلفه‌های مؤثر بر قیمت و سرانجام، پیش‌بینی قیمت آینده از جایگاهی بسیار مناسب برخوردار است. با توجه به نتایج به ‏دست‏ آمده، استفاده از الگوی پیشنهادی برای پیش‌بینی قیمت محصولات کشاورزی توصیه می‌شود.

کلیدواژه‌ها


عنوان مقاله [English]

Application of Hybrid Data Mining Model (Genetic Algorithm-Wavelet-Deep Neural Network-Monte Carlo Method) for Forecasting the Price of Agricultural Products: A Case Study of Future Price of Saffron in Agricultural Commodity Exchange

نویسندگان [English]

  • Reza Heydari 1
  • Seyed Mohammad Reza Haj Seyed Javady 2
1 Corresponding Author and Assistant Professor, Agricultural Planning, Economics and Rural Development Research Institute (APERDRI), Tehran, Iran
2 Researcher, Agricultural Planning, Economics and Rural Development Research Institute (APERDRI), Tehran, Iran
چکیده [English]

The price forecasting and its changes trend is one of the most important factors in decision making and formulating strategies related to agricultural products. This study aimed at presenting a hybrid data mining model for accurate price forecasting of agricultural products, including nonlinear models of wavelet transform, genetic algorithm, deep neural network and Monte Carlo technique. This proposed model involved a two-stage hybrid model and the base model of nonlinear-nonlinear. In this proposed model, the genetic algorithm for determining the optimal lag of price time series, the wavelet function for the de-noising of price data, the deep neural network for price forecasting, the Monte Carlo method for simulating the most probable price probability and finally, the complex soft calculations for "out-of-sample forecasting with new data set" were used. Results of comparison of the proposed model including "Genetic Algorithm-Wavelet Transform-Deep Neural Network-Monte Carlo", through evaluation criteria, with three competing models of "Genetic Algorithm-Deep Neural Network-Monte Carlo", "Genetic Algorithm-Wavelet Transform-Neural Network-Monte Carlo" and "Genetic Algorithm-Neural Network-Monte Carlo" showed that the proposed model had the better performance in forecasting of future price of saffron compared to the three competing models. Also, the use of deep neural network compared to neural network, the application of wavelet theory for de-noising and also the use of Monte Carlo technique to simulate the predicted prices, increase the forecasting accuracy of future price of saffron. In addition, the use of soft calculations showed that the proposed model had the necessary efficiency and high accuracy for short-term forecasting of the future price of saffron. Therefore, the present study has a good position in achieving the index of maximum accuracy, scenario making of future price trends, sensitivity analysis of components affecting the price and finally, forecasting the future price. Accordingly, the use of the proposed model to forecast the price of agricultural products is recommended.

کلیدواژه‌ها [English]

  • Forecasting
  • Future Price of Saffron
  • Genetic Algorithm
  • Wavelet Theory
  • Deep Learning Neural Network
  • Monte Carlo Method
Ajmera, R., Kook, N. and Crilley, J. (2012). Impact of commodity price movements on CPI inflation. Monthly Labor Review, 135: 29-43.
Alinejad, M., Bakhtiari, B. and Ghaderi, K. (2017). The comparison of Monte Carlo method and the combined method of fuzzy logic-PSO. Journal of Marine Engineering, 13(52): 105-112. (Persian)
Chen, Q., Lin, X., Zhong, Y. and Xie, Z. (2019). Price prediction of agricultural products based on wavelet analysis-LSTM. 2019 IEEE International Conference on Parallel and Distributed Processing with Applications, Big Data and Cloud Computing, Sustainable Computing and Communications, Social Computing and Networking (ISPA/BDCloud/SocialCom/SustainCom), Xiamen, China, pp. 984-990. DOI: 10.1109/ISPA-BDCloud-SustainCom-SocialCom48970.2019.00142.
Chuluunsaikhan, T., Ryu, G., Yoo, K.H., Rah, H. and Nasridinov, A. (2020). Incorporating deep learning and news topic modeling for forecasting pork prices: the case of South Korea. Journal of Agriculture, 10(11): 513. Available at file:///C:/Users/m.noushmand.APERI/Downloads/agriculture-10-00513-v2.pdf.
Ghias, M. (2014). Introduction on the method of Monte Carlo simulation. Scientific Journal of BASPARESH, 4(1): 67-77. (Persian)
Goodfellow, I., Bengio, Y. and Courville, A. (2016). Deep learning. NIT Press Book. Available at https://www. deeplearningbook.org.
Haven, E., Liu, X. and Shen, L. (2012). De-noising option prices with the wavelet method, European Journal of Operational Research, Elsevier, 222(1): 104-112. DOI: 10.1016/j.ejor.2012.04.020.
Jammazi, R. and Aloui, Ch. (2012). Crude oil price forecasting: experimental evidence from wavelet decomposition and neural network modeling. Energy Economics, 34: 828-841.
Li, X., He, K., Lai, K K. and Zou, Y. (2014). Forecasting crude oil price with multiscale de-noising ensemble model. Journal of Mathematical Problems in Engineering, (Special Issue): 1-19. Available at https://downloads.hindawi.com/journals/mpe/2014/716571.pdf.
Ly, R., Traore, F. and Dia, K. (2021). Forecasting commodity prices using long-short-term memory neural networks. IFPRI Discussion Paper 2000. Washington, DC: International Food Policy Research Institute (IFPRI). DOI: 10.2499/p15738coll2.134265.
Moghaddasi, R. and Jaleh Rajabi, M. (2011). Integrated modeling approach for the price forecasting of agricultural products. Journal of Economics and Development of Agriculture (Sciences and Industries of Agriculture), 5(3): 355-364. (Persian)
Mohammadi, T., Taklif, A. and Zamani, S. (2017). Natural gas price forecasting using the combination of wavelet transform and artificial neural network (case study: US market). Quarterly Journal of Iranian Economic Research, 22(71): 1-26. (Persian)
Nassar, L., Okwuchi, I.E., Saad, M., Karray, F. and Ponnambalam, K. (2020). Deep learning based approach for fresh produce market price prediction. Proceedings of IEEE 2020 International Joint Conference on Neural Networks (IJCNN), pp. 1-7.
Nazlioglu, S. and Soytas, U. (2011). World oil prices and agricultural commodity prices: evidence from an emerging market. Journal of Energy Economics, 33: 488-496.
Nazlioglu, S. (2011). World oil and agricultural commodity prices: evidence from nonlinear causality. Journal of Energy Policy, 39(5): 2935-2943.
Polyiam, K. and Boonrawd, P. (2017). A hybrid forecasting model of cassava price based on artificial neural network with support vector machine technique, Third International Conference on Information Management (ICIM), Chengdu, China, 2017, pp. 123-127. DOI: 10.1109/INFOMAN.2017.7950359.
Rai, R. and Mahmoudi Azar, M. (2014). The forecasting of the future return of the stock market using ARIMA, neural network and wavelet de-noising models. Quarterly Journal of Asset Management and Financing, 2(2:5): 1-16. (Persian)
Rai, R., Mohammadi, Sh. and Fenderski, H. (2015). The forecasting of the stock price index using neural network and wavelet transform. Quarterly Journal of Asset Management and Financing, 3(1:8): 55-74. (Persian)
Raikar, R.V., Wang, Ch-Yi., Shih, H.P. and Hong, J.H. (2016). Prediction of contraction scour using ANN and GA. Flow Measurement and Instrumentation, 50: 26-34.
Rasheed, A., Younis, M.S., Ahmad, F., Qadir, J. and Kashif, M. (2021). District wise price forecasting of wheat in Pakistan using deep learning. arXiv preprint arXiv:2103.04781. Available at https://arxiv.org/pdf/2103.04781.pdf.
Reboredo, J.C. and Rivera-Castro, M.A. (2013). A wavelet decomposition approach to crude oil price and exchange rate dependence. Economic Modelling, 32: 42-57. DOI: 10.1016/j.econmod.2012.12.028.
Sabu, K.M. and Kumar, T.M. (2020). Predictive analytics in agriculture: forecasting prices of Arecanuts in Kerala. Journal of Procedia Computer Science, 171: 699-708.
Sadeghi, H. and Dehghani Firoozabadi, Z. (2017). The de-noising of financial time series using wavelet analysis. Journal of Financial Engineering and Securities Management, 33: 299-315. (Persian)
Sangwana, K.S., Saxenaa, S. and Kanta, G. (2016). Optimization of machining parameters to minimize surface roughness using integrated ANN-GA approach, The 22nd CIRP Conference on Life Cycle Engineering, Procedia CIRP, 29: 305-310.
Shabri, A. and Samsudin, R. (2014). Daily crude oil price forecasting using hybridizing wavelet and artificial neural network model. Hindawi Publishing Corporation, Mathematical Problems Engineering, Vol. 2014, ID: 201402, pp. 10.
Shao, Y.E. and Dai, J.T. (2018). Integrated feature selection of ARIMA with computational intelligence approaches for food crop price prediction. Complexity, Vol. 2018, pp. 17. DOI: 10.1155/2018/1910520.
Shiferaw, Y.A. (2019). Time-varying correlation between agricultural commodity and energy price dynamics with Bayesian multivariate DCC-GARCH models. Physica A: Statistical Mechanics and its Applications, 526: 120807.
Vautard, R., Yiou, P. and Ghil, M. (1992). Singular-spectrum analysis: a toolkit for short, noisy chaotic signals. Physica D: Nonlinear Phenomena, 58(1-4): 95-126.
Wang, L. and Gupta, S. (2013). Neural networks and wavelet de-noising for stock trading and prediction. In: W. Pedrycz and S.M. Chen (Eds) Time series analysis, modeling and applications. Intelligent Systems Reference Library, vol. 47, pp. 229-247. Springer, Berlin, Heidelberg. Available at https://doi.org/10.1007/978-3-642-33439-9_11.
Wang, B., Liu, P., Chao, Z., Junmei, W., Chen, W., Cao, N., O’Hare, G.M.P. and Wen, F. (2018). Research on hybrid model of garlic short-term price forecasting based on big data. Journal of Computers, Materials and Continua (CMC), 57(2): 283-296.
Wang, D., Yue, Ch., Wei, Sh. and Lv, J. (2017). Performance analysis of four decomposition-ensemble models for one-day-ahead agricultural commodity futures price forecasting. Journal of Algorithms, 10(3): 108. DOI: 10.3390/a10030108.
Wang, L., Feng, J., Sui, X., Chu, X. and Mu, W. (2020). Agricultural product price forecasting methods: research advances and trend. British Food Journal, 122(7): 2121-2138.
Weng, Y., Wang, X., Hua, J., Wang, H., Kang, M. and Wang, F.Y. (2019). Forecasting horticultural products price using ARIMA model and neural network based on a large-scale data set collected by web crawler. Journal of IEEE Transactions on Computational Social Systems, 6(3): 547-553.
Xiong, T., Li, Ch., Bao, Y., Hu, Zh. and Zhang, L. (2015). A combination method for interval forecasting of agricultural commodity futures prices. Journal of Knowledge-Based Systems, 77: 92-102.
Yang, X. (2007). A modified particle swarm optimizer with dynamic adaptation Baoding: applied mathematics and Computation 51, June 2007, Elsevier, 189: 1205-1213.
Yang, Y., Chen, Y., Wang, Y., Li, C. and LiSchool, L. (2016). Modelling a combined method based on ANFIS and neural network improved by DE algorithm: a case study for short-term electricity demand forecasting. Journal of Applied Soft Computing, 49: 663-675.
Zhang, X., Lai, K.K. and Wang, Sh-Y. (2008). A new approach for crude oil price analysis based on Empirical Mode DecompositionEnergy Economics, Elsevier, 30(3): 905-918.
Zhang, D., Chen, S., Liwen, L. and Xia, Q. (2020). Forecasting agricultural commodity prices using model selection framework with time series features and forecast horizons. Journal of IEEE Access, 8: 28197-28209.
Zolfaghari, M., Sahabi, B. and Bakhtiaran, M.J. (2020). Designing a model for forecasting of return of the total stock market index (with emphasis on combined models of deep learning network and GARCH models). Quarterly Journal of Asset Management and Financing, 42: 138-171. (Persian)