摘要
基于模糊信息粒化和交叉验证算法的支持向量机(CV-SVM)预测时序回归模型,首先以欧洲碳排放配额(European union allowances,EUA)结算价为数据样本,预测未来连续5天的碳价波动情况,验证模型的可靠性及可用性.在得出科学结论的基础上,以我国碳交易市场中湖北碳排放配额(Hubei emission allowances,HBEA)结算价及北京碳排放配额(Beijing emission allowances,BEA)结算价,两组典型的价格数据为例,利用该模型进行训练,对我国碳交易市场中未来连续5天内碳价的变化趋势和波动区间给出有效预测.结果显示,该模型对两类碳交易市场的预测结果均较为理想,预测值误差率最大为3.36%,但针对欧洲碳交易市场的预测更为精确,预测值误差率不超过1.21%,在一定程度上反映了中国的碳交易市场尚不成熟,碳价波动规律性较弱,需要在正确的政策引导下快速、健康发展.
This paper proposed a support vector machine(SVM) time series regression predication model of carbon price based on fuzzy information granulation and cross validation algorithm. European Union allowances(EUA) settlement price is used as the data sample firstly, then the data will be trained to forecast the fluctuation of carbon price in the next five consecutive days, and verify the reliability and usability of the model. On the basis of the scientific conclusion, this paper makes use of two typical price data in China carbon trading market: Hubei emission allowances(HBEA) settlement price and Beijing emission allowances(BEA) settlement price as examples to train by this model and forecast the changing trend and fluctuation range of carbon price in the next five consecutive days. The results show that the model is ideal for two kinds of carbon trading market(prediction error rate is up to 3.36%), but the forecast for the European carbon trading market is more accurate(prediction error rate does not exceed 1.28%). The results to a certain extent reflect that China carbon trading market is not mature, carbon price fluctuation regularity is weak, and the market needs right policies to guide the rapid and healthy development.
引文
[ 1 ] 凤振华. 碳市场复杂系统价格波动机制与风险管理研究[D]. 合肥:中国科学技术大学,2012: 21-33.
[ 2 ] BENZ E, TRUCK S. Modeling the price dynamics of CO2 emission allowances [J]. Energy Economics, 2009, 31(1): 4-15.
[ 3 ] CONRAD C, RITTLER D, ROTFUB W. Modeling and explaining the dynamics of European Union allowance prices at the high-frequency[J]. Energy Economics, 2012, 34(1): 316-326. DOI:10.1016/j.eneco.2011.02.011.
[ 4 ] CHEVALLIER J, SEVI B. On the realized volatility of the ECX CO2emissions 2008 futures contract: distribution, dynamics and forecasting[J]. Annual Finance, 2011,7 (1): 1-29.
[ 5 ] SEGNON M, LUX T, GUPTA R. Modeling and forecasting the volatility of carbon dioxide emission allowance prices: a review and comparison of modern volatility models[J]. Renewable and Sustainable Energy Reviews, 2017,69:692-704.
[ 6 ] CHEVALLIER J. Nonparametric modeling of carbon prices[J]. Energy Economics, 2011, 33(6): 1267-1282.
[ 7 ] FAN Xinhua, LI Shasha, TIAN Lixin. Chaotic characteristic identification for carbon price and a multi-layer perception network prediction model[J]. Expert Systems with Applications, 2015 (42): 3945-3952.
[ 8 ] 孙睿, 况丹, 常冬青. 碳交易的“能源-经济-环境”影响及碳价合理区间测算[J].中国人口资源与环境, 2014,24(7): 82-90.SUN Rui, KUANG Dan, CHANG Dongqing. Analysis on carbon trading effects upon energy-economic-environment and calculation of reasonable carbon price intervals[J]. China Population Resources and Environment, 2014,24(7): 82-90.(in Chinese)
[ 9 ] FENG Zhenhua, ZOU Lele, WEI Yiming. Carbon price volatility: evidence from EU ETS[J]. Applied Energy, 2011,88(3): 590-598.
[10] ZHU Bangzhu, SHI Xuetao, CHEVALLIER J, et al. An adaptive multiscale ensemble learning paradigm for non-stationary and nonlinear energy price time series forecasting[J]. Journal of Forecasting, 2016, 35(7): 633-651.
[11] ZHU Bangzhu, WEI Yiming. Carbon price prediction with a hybrid ARIMA and least squares support vector machines methodology[J].Omega-International Journal of Management Science, 2013, 41(3):517-524.
[12] GUOBRANDSDOTTIR H N, HARALDSSON H O. Predicting the price of EU ETS carbon credits[J]. System Engineering Procedia,2011 (1): 481-489.
[13] 魏一鸣, 刘兰翠, 范英, 等. 中国能源报告(2008):碳排放研究[M]. 北京:科学出版社,2008.
[14] PAOLETTA M S, TASCHINI L. An econometric analysis of emission trading allowances[J]. Social Science Electronic Publishing, 2006, 32(10):2022-2032.
[15] 凤振华, 魏一鸣. 欧盟碳市场系统风险和预期收益的实证研究[J]. 管理科学, 2011, 8(3): 451-455.FENG Zhenhua, WEI Yiming. Systematic risk and expectations of returns in EU carbon market[J]. Chinese Journal of Management, 2011, 8(3): 451-455.(in Chinese)
[16] 张跃军, 魏一鸣. 化石能源市场对国际碳市场的动态影响实证研究[J]. 管理评论, 2010, 21(6): 34-41.ZHANG Yuejun, WEI Yiming. Interpreting the complex impact of fossil fuel markets on the EU ETS futures markets: an empirical evidence[J]. Management Review, 2010, 21(6): 34-41.(in Chinese)
[17] 张晨, 胡贝贝. 基于误差校正的多因素BP国际碳市场价格预测[J]. 价格月刊, 2017 (1): 11-18.ZHANG Chen, HU Beibei. Expectation of multiple factors BP international carbon market price based on error correction[J]. Prices Monthly, 2017 (1): 11-18. (in Chinese)
[18] THEOPHILOS P, PERIKLIS G, EFTHIMIOS S. Forecasting energy markets using support vector machines[J]. Energy Economics, 2014 (44): 135-142. DOI:10.1016/j.eneco.2014.03.017.
[19] 朱帮助, 魏一鸣. 基于GMDH-PSO-LSSVM的国际碳市场价格预测[J]. 系统工程理论与实践, 2011, 31(12): 2264-2271.ZHU Bangzhu, WEI Yiming. Carbon price predication based on integration of GMDH, particle swarm optimization and least squares support vector machines[J]. Systems Engineering-Theory & Practice, 2011, 31(12): 2264-2271.(in Chinese)
[20] 王国胤, 张清华, 胡军. 粒计算研究综述[J]. 智能系统学报, 2007 (6): 8-26.WANG Guoyin, ZHANG Qinghua, HU Jun. An overview of granular computing[J]. CAAI Transactions on Intelligent Systems, 2007 (6): 8-26. (in Chinese)
[21] ZADEH L A. Fussy Sets[J]. Information and Control, 1965 (8): 338-353.
[22] 喻胜华, 肖雨峰. 基于信息粒化和支持向量机的股票价格预测[J]. 财经理论与实践, 2011,32(6): 44-47.YU Shenghua, XIAO Yufeng. Forecasting method of stock price based on information granulation and support vector machine[J]. The Theory and Practice of Finance and Economic, 2011,32(6): 44-47.(in Chinese)
[23] PEDRYCZ W. Granular computing: an introduction[M].New York:Kluwer Academic Publisher,2003.
[24] 彭勇,陈俞强. 基于信息粒化的 SVM 时序回归预测[J]. 计算机系统应用,2013, 22 (5): 163-167.PENG Yong, CHEN Yuqiang. Time series regression and prediction based on information granulation and SVM[J]. Computer Systems & Applications, 2013, 22 (5): 163-167. (in Chinese)
[25] ZHU Bangzhu, HAN Dong, WANG Ping, et al. Forecasting carbon price using empirical mode decomposition and evolutionary least squares support vector regression[J]. Applied Energy, 2017,191: 521-530.
[26] 陈珂, 柯文德. 支持向量机的石油价格预测[J]. 计算机仿真,2011, 28(12): 327-330.CHEN Ke, KE Wende. Oil price prediction based on support vector machine[J]. Computer Simulation, 2011,28(12): 327-330. (in Chinese)