基于MVMD-BO-BiMamba的多因素棉花期货价格预测方法

doi:10.14088/j.cnki.issn0439-8114.2026.07.029

湖北农业科学 ›› 2026, Vol. 65 ›› Issue (7): 184-193.doi: 10.14088/j.cnki.issn0439-8114.2026.07.029

基于MVMD-BO-BiMamba的多因素棉花期货价格预测方法

陈聪^a,b, 朱静^{a b,c}, 苑孟于^a,b, 王硕超^a,b

新疆农业大学,a.计算机与信息工程学院; b.新疆农业信息化工程技术研究中心; c.智能农业教育部工程研究中心,乌鲁木齐 830052

收稿日期:2026-04-27 出版日期:2026-07-25 发布日期:2026-07-23
通讯作者: 朱静（1979-）,女,广西罗城人,副教授,博士,主要从事计算机应用研究,（电子信箱）Zhujing@xjau.edu.cn。
作者简介:陈聪（2000-）,男,四川遂宁人,在读硕士研究生,研究方向为农业经济时序预测,（电子信箱）320243416@stu.xjau.edu.cn。
基金资助:
科技创新2030—“新一代人工智能”重大项目（2022ZD0115805）; 新疆维吾尔自治区重大科技专项（2022A02011）; 新疆维吾尔自治区高校基本科研业务费科研项目（XJEDU2026J053）

A multi-factor cotton futures price prediction method based on MVMD-BO-BiMamba

CHEN Cong^a,b, ZHU Jing^a,b,c, YUAN Meng-yu^a,b, WANG Shuo-chao^a,b

a. College of Computer and Information Engineering; b. Xinjiang Agricultural Informatization Engineering Technology Research Center; c. Engineering Research Center of Intelligent Agriculture, Ministry of Education, Xinjiang Agricultural University,Urumqi 830052, China

Received:2026-04-27 Published:2026-07-25 Online:2026-07-23

摘要/Abstract

摘要： 针对棉花期货价格非线性、非平稳特性及多因素耦合带来的预测难题,以棉花期货结算价作为预测目标变量,提出融合多变量变分模态分解（MVMD）、贝叶斯优化（BO）与双向Mamba的预测模型。首先利用正交投影最大信息系数筛选关键特征;采用MVMD对多维特征序列进行联合分解,提取协同特征并抑制非平稳噪声。然后通过BO算法优化双向Mamba模型超参数以避免人工调参的盲目性。最后,将优化后的模型应用于各分量的双向时序依赖预测。多因素场景下的对比试验表明,该模型的均方根误差（RMSE）、平均绝对误差（MAE）和平均绝对百分比误差（MAPE）分别降至58.05元/t、46.04元/t和0.34%,决定系数（R²）提升至0.978,预测精度优于MVMD-BiMamba、BO-BiMamba、BiMamba、VMD-BO-BiMamba、Mamba、Transformer、LSTM基准模型。

关键词: 多变量变分模态分解（MVMD）, 贝叶斯优化（BO）, 双向Mamba, 结算价, 多因素, 棉花期货价格, 预测方法

Abstract: Aiming at the prediction difficulties caused by the nonlinear and non-stationary characteristics of cotton futures prices and multi-factor coupling, a prediction model integrating multivariate variational mode decomposition (MVMD), Bayesian optimization (BO), and bidirectional Mamba was proposed, with the cotton futures settlement price as the prediction target variable. First, the orthogonal projection maximum information coefficient was used to screen key features; MVMD was adopted to jointly decompose the multi-dimensional feature sequences, extracting collaborative features and suppressing non-stationary noise. Then, the hyperparameters of the bidirectional Mamba model were optimized by the BO algorithm to avoid the blindness of manual parameter tuning.Finally, the optimized model was applied to the bidirectional temporal dependency prediction of each component. The results showed that in the multi-factor scenario, the comparative experiment indicated that the root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE) of the model were reduced to 58.05 yuan/t, 46.04 yuan/t, and 0.34%, respectively, and the coefficient of determination (R²) increased to 0.978, with prediction accuracy superior to that of the MVMD-BiMamba, BO-BiMamba, BiMamba, VMD-BO-BiMamba, Mamba, Transformer, and LSTM benchmark models.

Key words: multivariate variational mode decomposition (MVMD), Bayesian optimization (BO), bidirectional Mamba, settlement price, multi-factor, cotton futures price, prediction method

中图分类号:

TP183

陈聪, 朱静, 苑孟于, 王硕超. 基于MVMD-BO-BiMamba的多因素棉花期货价格预测方法[J]. 湖北农业科学, 2026, 65(7): 184-193.

CHEN Cong, ZHU Jing, YUAN Meng-yu, WANG Shuo-chao. A multi-factor cotton futures price prediction method based on MVMD-BO-BiMamba[J]. HUBEI AGRICULTURAL SCIENCES, 2026, 65(7): 184-193.

参考文献

[1] 蔡咏梅,王一帆,张创奇,等.我国棉花期货价格发现及波动溢出效应研究[J].中国证券期货,2025(2):27-35.
[2] 王莉珊. 中国棉花期货价格波动及风险测度研究[D].河北保定:河北农业大学,2023.
[3] 冯雪倩,肖杨.中国棉花产业链价格传导机制浅析[J].合作经济与科技,2024(9):62-65.
[4] 张玉静. 中国棉花期货价格影响因素分析[D].上海:上海财经大学,2020.
[5] 何嵘. 我国农产品期货价格波动影响因素研究[D].辽宁大连:大连理工大学,2024.
[6] LI X, WANG Y.Nonlinear and non-stationary characteristics of cotton futures prices and their impact on forecasting performance[J]. Journal of forecasting, 2023, 42(3): 567-585.
[7] 原云霄,于惠兰,崔静.基于ARMA与GARCH-M模型对我国豆粕期货价格波动的分析预测[J].饲料博览,2021(2):47-54.
[8] 陈新华,刘洁.基于ARIMA模型的豆粕期货价格预测方法研究及启示[J].南方农村,2020,36(2):23-26.
[9] 范俊明,刘洪久,胡彦蓉.基于LSTM深度学习的大豆期货价格预测[J].价格月刊,2021(2):7-15.
[10] 杨学威. 基于机器学习算法的股指期货价格预测模型研究[J].软件工程,2022,25(12):1-8.
[11] 陈立平,邢小丹,张玉亭,等.基于LSTM的红枣期货价格预测方法[J].农业与技术,2024,44(1):162-165.
[12] SHI Z W. Mamba Stock: Selective state space model for stock prediction[EB/OL].(2024-02-29).https://arxiv.org/abs/2402.18959.
[13] GHANBARI E, AVAR A.Short-term wind power forecasting using the hybrid model of multivariate variational mode decomposition (MVMD) and long short-term memory (LSTM) neural networks[J]. Electrical engineering, 2025, 107(3): 2903-2933.
[14] 邢蕾,林思扬.基于改进VMD和RBF的股票预测研究[J].长春工业大学学报,2024,45(2):164-171.
[15] 王安,赵德彦,宋瑞军,等.基于精确负荷预测的新能源发电与储能调控技术[J].信息技术,2025(10):166-170,176.
[16] ZHANG X Q,REN H, LIU J W, et al.A monthly temperature prediction based on the CEEMDAN-BO-BiLSTM coupled model[J].Scientific reports, 2024, 14: 808.
[17] CAO T N, CUI Y F, TAN H T, et al.A field verification denoising method for partial discharge ultrasonic sensors based on IPSO-optimated multivariate variational mode decomposition combined with improved wavelet transforms[J]. Sensors, 2025, 25(24):7506.
[18] CIHAN P.Bayesian hyperparameter optimization of machine learning models for predicting biomass gasification gases[J]. Applied sciences, 2025, 15(3):1018.
[19] GU A, DAO T. Mamba: Linear-time sequence modeling with selective state spaces[EB/OL].(2024-05-31). https://arxiv.org/abs/2312.00752.

基于MVMD-BO-BiMamba的多因素棉花期货价格预测方法

A multi-factor cotton futures price prediction method based on MVMD-BO-BiMamba

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	万新洋, 史铁丑, 冯文钊. 基于层次分析法和多因素综合评价法的沧州市土地集约利用评价[J]. 湖北农业科学, 2026, 65(6): 253-257.
[2]	李志英, 汪琳, 龙晔, 陈宁歆, 李晨晨. 昆明呈贡区高校土地利用集约度评价研究[J]. 湖北农业科学, 2020, 59(20): 177-182.
[3]	张转, 卫新东, 王筛妮, 郭树延. 渭北旱塬区耕地质量等别监测样点布控研究[J]. 湖北农业科学, 2018, 57(2): 48-54.