首页|基于VMD-CSSA-LSTM组合模型的股票价格预测

基于VMD-CSSA-LSTM组合模型的股票价格预测

扫码查看
原文链接
  • 万方数据
  • 维普
针对股票价格非平稳、非线性和高复杂等特性引发的预测难度大的问题,建立一种基于变分模态分解(Variational Mode Decomposition,VMD)-Circle混沌映射的麻雀搜索算法(Circle Sparrow Search Algorithm,CSSA)-长短期记忆(Long Short-Term Memory,LSTM)神经网络的组合模型——VMD-CSSA-LSTM.首先,利用VMD将原始股票收盘价数据分解为若干本征模态函数(Intrinsic Mode Func-tion,IMF)分量.然后,采用Circle混沌映射的SSA算法对LSTM神经网络的隐含层神经元、迭代次数、学习率进行优化,将最优参数拟合至 LSTM 网络中.最后,对每个IMF分量建模预测,将各分量预测结果叠加得到最终结果.实验结果表明,与其他模型相比,本文模型在多支股票数据集上的均方根误差(RMSE)、平均绝对误差(MAE)及平均绝对百分比误差(MAPE)均达到最小,预测股票收盘价格误差在 0 附近波动,稳定性更优、拟合更佳、精确度更高.
Stock price prediction based on VMD-CSSA-LSTM combination model
To address the problems of stock price prediction due to its non-static,highly complex and random fluc-tuations,a combination model based on Variational Mode Decomposition(VMD)-Circle Sparrow Search Algorithm(CSSA)-Long Short-Term Memory(LSTM)neural network is established.The original stock closing data is decom-posed into several Intrinsic Mode Function(IMF)components by VMD,and then the CSSA is used to optimize the parameters of hidden layer neurons,iteration number and learning rate of LSTM,and the optimal parameters are fit-ted into the LSTM,where each IMF component is modeled and predicted,and the prediction results of IMF compo-nent are superimposed to obtain the final result.Experiments show that the RMSE,MAE and MAPE of the proposed model are minimized on multiple stock datasets,the error of the predictied closing prices of individual stocks fluctu-ates around 0,which is more stable with better fitting and higher accuracy.

stock price forecastingvariational mode decomposition(VMD)sparrow search algorithm(SSA)Circle chaos mappinglong short-term memory(LSTM)

黄后菊、李波

展开 >

辽宁工业大学 电子与信息工程学院,锦州,121001

股票价格预测 变分模态分解 麻雀搜索算法 Circle混沌映射 长短期记忆网络

辽宁省教育厅基本科研项目(面上项目)国家自然科学基金辽宁省自然科学基金

JYTM20230862516791162020-MS-292

2024

10.13878/j.cnki.jnuist.20230903002

南京信息工程大学学报
南京信息工程大学

南京信息工程大学学报

CSTPCD北大核心
影响因子：0.737
ISSN：1674-7070
年,卷(期)：2024.16(3)