To Rise or to Fall? Chinese Medicinal Materials Price Index Trend Prediction using GA-XGBoost Feature Selection and Bi-GRU Deep Learning

Abstract

As the national Chinese medicine market develops, Chinese medicinal materials price index (CMMPI) trend is worthy of attention. Predicting future CMMPI trend plays a significant role in risk prevention, cultivation, and trade for farmers and investors. This study aims to design a high-precision model to predict the future trend of the CMMPI. The model incorporates environmental factors such as weather conditions and air quality that have a greater impact on the growth of Chinese medical plants and the supply of Chinese medicinal materials market. Specifically, we collected multi-source heterogeneous data, including weather data, air quality data, and historical CMMPI data, to construct informative features. Additionally, we proposed a feature selection method based on the genetic algorithm and XGBoost to select features. Finally, we transferred the selected features to the bidirectional GRU deep learning to realize the accurate prediction of the CMMPI trend. We collected 46 CMMPI datasets to test the proposed model. The results show that the proposed model obtained more superior prediction compared to the state-of-the-art methods, and specialized in predicting long-term goal (90 days). Taking the Yunnan and Tibet origin index as examples, the experiment results also show the weather and air quality data can improve the prediction performance, as these factors are known to influence the growth and market supply of Chinese medicinal materials.

Keywords

Chinese medicinal material price index / genetic algorithm / XGBoost / feature selection / deep learning / prediction

References

[1] Aljohani H M, Elhag A A. Using statistical model to study the daily closing price index in the Kingdom of Saudi Arabia (KSA). Complexity, 2021, 2021: 1-5.

[2] Bisbis M B, Gruda N, Blanke M. Potential impacts of climate change on vegetable production and product quality–A review. Journal of Cleaner Production, 2018, 170: 1602-1620.

[3] Brown J N, Ash A, MacLeod N. et al.. Diagnosing the weather and climate features that influence pasture growth in Northern Australia. Climate Risk Management, 2019, 24: 1-12.

[4] Canito J, Ramos P, Moro S. et al.. Unfolding the relations between companies and technologies under the Big Data umbrella. Computers in Industry, 2018, 99: 1-8.

[5] Chan K. Chinese medicinal materials and their interface with western medical concepts. Journal of Ethnopharmacology, 2005, 96(1–2): 1-18.

[6] Chang F, Mao Y D. Study on early-warning of Chinese materia medica price base on ARMA model. China Journal of Chinese Materia Medica, 2014, 39(9): 1721-1723

[7] Chen Y, Lei L, Bi Y. et al.. Quality control of Glehniae Radix, the Root of Glehnia Littoralis Fr. Schmidt ex Miq., along its value chains. Frontiers in Pharmacology, 2021, 12: 729554.

[8] Chi X, Zhang Z, Xu X. et al.. Threatened medicinal plants in China: Distributions and conservation priorities. Biological Conservation, 2017, 210: 89-95.

[9] Chung H, Shin K. Genetic algorithm-optimized multi-channel convolutional neural network for stock market prediction. Neural Computing and Applications, 2020, 32(12): 7897-7914.

[10] Cunningham A B, Long X. Linking resource supplies and price drivers: Lessons from Traditional Chinese Medicine (TCM) price volatility and change, 2002–2017. Journal of Ethnopharmacology, 2019, 229: 205-214.

[11] El-Rashidy M A. A novel system for fast and accurate decisions of gold-stock markets in the short-term prediction. Neural Computing and Applications, 2021, 33(1): 393-407.

[12] Fang W, Zhang S, Xu C. Improving prediction efficiency of Chinese stock index futures intraday price by VIX-Lasso-GRU Model. Expert Systems with Applications, 2024, 238: 121968.

[13] Fernández A, García S, Herrera F. et al.. SMOTE for learning from imbalanced data: Progress and challenges, marking the 15-year anniversary. Journal of Artificial Intelligence Research, 2018, 61: 863-905.

[14] Frühwirth M, Sögner L. Weather and SAD related mood effects on the financial market. The Quarterly Review of Economics and Finance, 2015, 57: 11-31.

[15] Gallic E, Vermandel G. Weather shocks. European Economic Review, 2020, 124: 103409.

[16] Gupta A, Singh P P, Singh P, et al. (2019). Medicinal plants under climate change: Impacts on pharmaceutical properties of plants. Climate Change and Agricultural Ecosystems: 181–209.

[17] Gupta U, Bhattacharjee V, Bishnu P S. StockNet-GRU based stock index prediction. Expert Systems with Applications, 2022, 207: 117986.

[18] Han N, Qiao S, Yuan G. et al.. A novel Chinese herbal medicine clustering algorithm via artificial bee colony optimization. Artificial Intelligence in Medicine, 2019, 101: 101760.

[19] Hu R, Lai K, Luo B, Tang R J, Huang R B, Ye X X. The medicinal plant used in the Guangxi Fangcheng Golden Camellias national nature reserve, a coastal region in southern China. Journal of Ethnobiology and Ethnomedicine, 2023, 19(1): 32.

[20] Huber J, Müller S, Fleischmann M. et al.. A data-driven newsvendor problem: From data to decision. European Journal of Operational Research, 2019, 278(3): 904-915.

[21] Huck N. Large data sets and machine learning: Applications to statistical arbitrage. European Journal of Operational Research, 2019, 278(1): 330-342.

[22] Jørgensen S L, Termansen M, Pascual U. Natural insurance as condition for market insurance: Climate change adaptation in agriculture. Ecological Economics, 2020, 169: 106489.

[23] Katal A, Wazid M, Goudar R H. Big data: Issues, challenges, tools and good practices. Proceedings of the Sixth International Conference on Contemporary Computing (IC3), 2013August 08–10, 2013

[24] Khan A, Vibhute A D, Mali S. et al.. A systematic review on hyperspectral imaging technology with a machine and deep learning methodology for agricultural applications. Ecological Informatics, 2022, 69: 101678.

[25] Lazo J K, Lawson M, Larsen P H. et al.. US economic sensitivity to weather variability. Bulletin of the American Meteorological Society, 2011, 92(6): 709-720.

[26] Le Gouis J, Oury F X, Charmet G. How changes in climate and agricultural practices influenced wheat production in Western Europe. Journal of Cereal Science, 2020, 93: 102960.

[27] Li F, Song Q, Chen C. et al.. Forecast of Panax notoginseng price index based on LSTM neural network. Modern Chinese Medicine, 2014, 21(4): 536-541

[28] Li J, Cheng K, Wang S. et al.. Feature selection: A data perspective. ACM Computing Surveys (CSUR), 2017, 50(6): 1-45.

[29] Li Y, Yang H, Liu J, Fu H, Chen S. Predicting price index of Chinese herval medicines in China. Journal of Huazhong Agricultural University, 2021, 40(6): 50-59

[30] Long W, Gao J, Bai K. et al.. A hybrid model for stock price prediction based on multi-view heterogeneous data. Financial Innovation, 2024, 10(1): 48.

[31] Lu H, Mazumder R. Randomized gradient boosting machine. SIAM Journal on Optimization, 2020, 30(4): 2780-2808.

[32] Ma G, Ma D, Shao X. On price forecasting of radix Notoginseng based on genetic BP neural network. Journal of Tianjin Normal University (Natural Science Edition), 2019, 7(6): 76-80

[33] Maldonado S, Vairetti C, Fernandez A. et al.. FWSMOTE: A feature-weighted oversampling approach for unbalanced classification. Pattern Recognition, 2020, 124: 108511.

[34] Mao Y, Chang F. Prediction of Chinese herbal medicine price index based on gray GM(1,1) prediction model. China Pharmacy, 2014, 25(23): 2200-2202

[35] Parvandeh S, Yeh H W, Paulus M P. et al.. Consensus features nested cross-validation. Bioinformatics, 2020, 36(10): 3093-3098.

[36] Severen C, Costello C, Deschenes O. A Forward-Looking Ricardian Approach: Do land markets capitalize climate change forecasts?. Journal of Environmental Economics and Management, 2018, 89: 235-254.

[37] Shahzad F. Does weather influence investor behavior, stock returns, and volatility? Evidence from the Greater China region. Physica A: Statistical Mechanics and its Applications, 2019, 523: 525-543.

[38] Shrivastav L K, Kumar R. An ensemble of random forest gradient boosting machine and deep learning methods for stock price prediction. Journal of Information Technology Research, 2022, 15(1): 1-19

[39] Wang W Y, Xie Y, Zhou H. et al.. Contribution of traditional Chinese medicine to the treatment of COVID-19. Phytomedicine, 2021, 85: 153279.

[40] Yang C Y, Jhang L J, Chang C C. Do investor sentiment, weather and catastrophe effects improve hedging performance? Evidence from the Taiwan options market. Pacific-Basin Finance Journal, 2016, 37: 35-51.

[41] Yao Q, Huang Y, Lu D. et al.. Study on price index fluctuation of China’s authentic medicinal materials-taking Guizhou Ainaxiang, Uncaria and other eight high-quality traditional Chinese medicinal meterials as examples. Price: Theory & Practice, 2021, 2: 87-90

[42] Yin L, Li B, Li P. et al.. Research on stock trend prediction method based on optimized random forest. CAAI Transactions on Intelligence Technology, 2023, 8(1): 274-284.

[43] Yu M, Guo C. Using news to predict Chinese medicinal material price index movements. Industrial Management & Data Systems, 2018, 118(5): 998-1017.

[44] Yu T, Li J, Yu Q. et al.. Knowledge graph for TCM health preservation: Design, construction, and applications. Artificial Intelligence in Medicine, 2017, 77: 48-52.

[45] Zhang Q, Zhang Y, Bao F. et al.. Incorporating stock prices and text for stock movement prediction based on information fusion. Engineering Applications of Artificial Intelligence, 2024, 127: 107377.

[46] Zhong L, Hu L, Zhou H. Deep learning based multi-temporal crop classification. Remote Sensing of Environment, 2019, 221: 430-443.

RIGHTS & PERMISSIONS

Systems Engineering Society of China and Springer-Verlag GmbH Germany