论文部分内容阅读
碳排放交易市场中不同期货价格波动及其相互影响较为复杂,价格趋势预测也在金融投资领域占有重要地位。针对碳交易市场中非线性预测问题,选取欧盟配额期货与碳排放核证减排量期货相关参数作为研究对象,运用协整关系检验确定其是否具有长期协整关系,采用Granger因果检验确定其领先滞后关系,将具有领先关系的期货参数作为部分输入变量,建立遗传算法改进的运用不同小波函数的神经网络模型,对具有滞后关系的期货价格趋势进行预测,并与改进前的BP小波神经网络模型预测结果进行对比。实验结果表明,碳排放交易市场中期货价格之间存在长期均衡协整关系,改进的模型可以有效刻画期货价格序列变化趋势,为碳排放交易提供良好的投资建议。
Fluctuations in the prices of different futures in the carbon emissions trading market and their interactions are complicated, and price trend forecasts also play an important role in the financial investment field. Aiming at the problem of non-linear forecasting in carbon market, this paper chooses the relevant parameters of the EU quota futures and the carbon emission reduction futures as the research object, and uses the cointegration test to determine if it has a long-term co-integration relationship and adopts Granger causality test to determine its lead Lagged relationship, the futures parameters with the leading relationship are taken as a part of the input variables, and the improved neural network model with different wavelet functions is established by using genetic algorithm to forecast the futures price trend with lagged relationship and compared with the pre-modified BP wavelet neural network model The forecast results are compared. The experimental results show that there is a long-term equilibrium co-integration relationship between the futures prices in the carbon emissions trading market. The improved model can effectively characterize the trend of the futures price series and provide good investment advice for carbon emissions trading.