【摘要】 针对支持向量回归(SVR)模型参数选择困难以及在碳期货价格预测中模型误差高的问题，提出一种基于改进粒子群算法-支持向量回归(TSA-PSO-SVR)的期货价格预测模型.通过改进粒子群算法惯性权重实现局部搜索和全局搜索能力的平衡，引入被囊群算法(TSA)对粒子群位置更新公式进行优化，利用改进的粒子群算法(TSA-PSO)找出最优参数有效解决支持向量回归参数选择盲目性的问题；将得到的最优参数应用于期货价格预测模型.选取福建碳交易市场的碳交易价格进行预测，与支持向量回归(SVR)、差分自回归移动平均模型(ARIMA)、长短期记忆模型(LSTM)模型作对比，实验结果表明TSA-PSO-SVR模型有效克服了高预测误差和参数选择随机性的问题，并具有较高的泛化能力.更多还原

【Abstract】 To solve the problem of difficult parameter selection of support vector regression(SVR) model and high error in carbon futures price prediction, a futures price prediction model based on improved particle swarm optimization algorithm support vector regression(TSA-PSO-SVR) was proposed. The inertia weight of particle swarm optimization(PSO) algorithm was improved to balance the local search and global search ability. The trapped swarm optimization(TSA) algorithm was introduced to optimize the PSO position updating formula. TSA-PSO algorithm was used to find the optimal parameters to effectively solve the blind selection of support vector regression parameters. The optimal parameters are applied to the futures price prediction model. Compared with SVR, ARIMA and LSTM models, the experimental results show that TSA-PSO-SVR model effectively overcomes the problems of high prediction error and randomness of parameter selection. And it has high generalization ability.更多还原