碳排放权价格均值回归的周期及振幅
|
摘要
本文运用谱估计技术分析了欧盟碳排放权价格均值回归周期、幅度及其与WTI,PMI之间的耦合关系.研究表明:1)EUA现货价格具有显著的均值回归周期振荡特征,周期约在15.5个月与3个月之间;振幅约在-2.298到4.823之间;2)EUA现货价格均值回归与WTI原油价格指数的耦合周期在3个月到12个月之间,耦合振幅在0.1958到0.8843之间,与PMI指数耦合周期约为4个月到11个月之间.耦合振幅在0.1652到2.134之间;3)在所有耦合周期模态下,耦合周期越长,耦合振幅越小.
This paper analysis of the price of EU carbon emissions mean return period, amplitude and coupling relations through the spectrum estimation technique. The results show that: 1) EUA spot prices have significant mean reversion characteristics of periodic oscillation, cycle between about 15.5 and 3 months; the amplitude between-2.298 to 4.823;2) Coupling cycle of EUA spot price mean reversion and WTI crude oil price index in 3 to 12 months, coupled amplitudes between 0.1958 to 0.8843, and the PMI index in 4 to 11 months. coupling amplitudes between 0.1652 to 2.134. The amplitude 3) in all the coupling cycle mode. The coupling cycle is long, the smaller the coupling amplitude.
This paper analysis of the price of EU carbon emissions mean return period, amplitude and coupling relations through the spectrum estimation technique. The results show that: 1) EUA spot prices have significant mean reversion characteristics of periodic oscillation, cycle between about 15.5 and 3 months; the amplitude between-2.298 to 4.823;2) Coupling cycle of EUA spot price mean reversion and WTI crude oil price index in 3 to 12 months, coupled amplitudes between 0.1958 to 0.8843, and the PMI index in 4 to 11 months. coupling amplitudes between 0.1652 to 2.134. The amplitude 3) in all the coupling cycle mode. The coupling cycle is long, the smaller the coupling amplitude.
引文
[1]SONG Yuchen.A summary of studies on the theory of stock price mean reversion and its quantitative analysis method[J].Taxation and Economy,2006,(1):73–74.)(宋玉臣.股票价格均值回归理论研究综述[J].税务与经济,2006,(1):73–74.)
[2]ROSSEN A,GOLOSNOY V.Modeling dynamics of metal price series via state space approach with two common factors[J].Empirical Economics,2014:1–25.
[3]CASTROV.The portuguese stock market cycle[J].Oecd Journal Journal of Business Cycle Measurement&Analysis,2013,2013(1):1–23.
[4]HSIEHH.Do managers of global equity funds outperform their respective style benchmarks?an empirical investigation[J].International Business&Economics Research Journal,2012,11(3):269–274.
[5]NACCACHE T.Oil price cycles and wavelets[J].Energy Economics,2011,33(2):338–352.
[6]PIAZZESI M,SCHNEIDER M.Trend and cycle in bond premia[J].Staff Report,2009,40(5):839–843.
[7]NOBRE C,JR R B,COSTA A G.Biospeckle laser spectral analysis under inertia moment,entropy and cross-spectrum methods[J].Optics Communications,2009,282(11):2236–2242.
[8]BLANTER E,MOUEL J,SHNIRMAN M,et al.Kuramoto model with non-symmetric coupling reconstructs variations of the solarcycle period[J].Solar Physics,2016,291(3):1003–1023.
[9]KLINGAMAN N P,WOOLNOUGH S J.The role of air–sea coupling in the simulation of the Madden–Julian oscillation in the Hadley centre model[J].Quarterly Journal of the Royal Meteorological Society,2014,140(684):2272–2286.
[10]QIAN Youhua,ZHANG Wei.Periodic solutions for coupled van del pol oscillatorys of two-degrees-of-freedom by homotopy analysis method[J].Science&Technology Review,2008,26(22):21–25.(钱有华,张伟.两自由度耦合van del Pol振子周期解的同伦分析方法[J].科技导报,2008,26(22):21–25.)
[11]MOHAMMED I,SELJAK U.Analytic model for the matter power spectrum,its covariance matrix and baryonic effects[J].Monthly Notices of the Royal Astronomical Society,2017,445(4):3382–3400.
[12]JACKEL B.Characterization of auroral radar power spectra and autocorrelation functions[J].Radio Science,2016,35(4):1009–1023.
[13]KONDRASHOV D,SHPRITS Y,GHIL M.Gap filling of solar wind data by singular spectrum analysis[J].Geophysical Research Letters,2015,37(37):78–82.
[14]LU Chenglong,KUANG Cuilin,YI Zhonghai,et al.Singular spectrum analysis filter method for mitigation of GPS multipath error[J].Geomatics and Information Science of Wuhan University,2015,40(7):924–931.(卢辰龙,匡翠林,易重海,等.奇异谱分析滤波法在消除GPS多路径中的应用[J].武汉大学学报(信息科学版),2015,40(7):924–931.)
[15]YUE Shun,LI Xiaoqi,ZHAI Changzhi.Determining the singular spectrum embedding dimension based on an improved Cao algorithm[J].Engineering of Surveying and Mapping,2015,24(3):64–68.(岳顺,李小奇,翟长治.基于改进Cao算法确定奇异谱嵌入维数及应用[J].测绘工程,2015,24(3):64–68.)
[16]YOU Fengchun,DING Yuguo,ZHOU Yu,et al.Application of SVD and SSCA in the prediction of summer precipitation diagnosisin North China[J].Journal of Applied Meteorological Science,2003,14(2):176–187.(尤凤春,丁裕国,周煜,等.奇异值分解和奇异交叉谱分析方法在华北夏季降水诊断中的应用[J].应用气象学,2003,14(2):176–187.)
[17]BENOWITZ B,SHIELD M,DEODATIS G.Determining evolutionary spectra from non-stationary autocorrelation functions[J].Probabilistic Engineering Mechanics,2015,41:73–88.
[18]NI Lili,ZHANG Yanlan.The determination of the best embedding dimension of bridge deformation time series based on G--P method and Cao method[J].Northern Communications,2017,(5):5–7.(泥立丽,张艳兰.基于G--P法和Cao法的桥梁变形时间序列最佳嵌入维数的确定[J].北方交通,2017,(5):5–7.)
[19]GU C,ROHLING J,LIANG X,et al.Impact of dispersed coupling strength on the free running periods of circadian rhythms[J].Physical Review E,2016,93(3-1):032414.
[20]WU Yilin,SHEN Zhiping.The tracking problem in networked systems with periodic signal reference input[J].Control Theory&Applications,2016,33(5):685–693.(邬依林,沈志萍.网络化系统周期信号的跟踪[J].控制理论与应用,2016,33(5):685–693.)
[21]WANG Lei,XU Genqi.Stabilization of the complex wave network with two circuits[J].Control Theory&Applications,2011,28(6):759–762.(王雷,许跟起.双回路复杂杂波网络的镇定[J].控制理论与应用,2011,28(6):759–762.)
1由于已证实第1阶段并不存在均值回归,故未选用第1阶段数据.
2 指欧洲排放权配额.
3 经典谱估计又分为布莱克曼–杜基谱估计器,Blackman和J.Tukey,BT谱估计器)(间接法)和周期图法(直接法);现代谱估计可分为参数模型谱估计和非参数模型谱估计.
4 Tukey根据Wiener-Khintchine定理提出了对有限长数据进行谱估计的自相关法,即利用有限长度数据估计自相关函数,再对该自相关函数求傅立叶变换,从而得到谱的估计.
5月度数据是反映时间序列中长周期的依据.
6特征值分别表示信号不同阶数的主成分,特征值较小的成分反映噪声项.
[2]ROSSEN A,GOLOSNOY V.Modeling dynamics of metal price series via state space approach with two common factors[J].Empirical Economics,2014:1–25.
[3]CASTROV.The portuguese stock market cycle[J].Oecd Journal Journal of Business Cycle Measurement&Analysis,2013,2013(1):1–23.
[4]HSIEHH.Do managers of global equity funds outperform their respective style benchmarks?an empirical investigation[J].International Business&Economics Research Journal,2012,11(3):269–274.
[5]NACCACHE T.Oil price cycles and wavelets[J].Energy Economics,2011,33(2):338–352.
[6]PIAZZESI M,SCHNEIDER M.Trend and cycle in bond premia[J].Staff Report,2009,40(5):839–843.
[7]NOBRE C,JR R B,COSTA A G.Biospeckle laser spectral analysis under inertia moment,entropy and cross-spectrum methods[J].Optics Communications,2009,282(11):2236–2242.
[8]BLANTER E,MOUEL J,SHNIRMAN M,et al.Kuramoto model with non-symmetric coupling reconstructs variations of the solarcycle period[J].Solar Physics,2016,291(3):1003–1023.
[9]KLINGAMAN N P,WOOLNOUGH S J.The role of air–sea coupling in the simulation of the Madden–Julian oscillation in the Hadley centre model[J].Quarterly Journal of the Royal Meteorological Society,2014,140(684):2272–2286.
[10]QIAN Youhua,ZHANG Wei.Periodic solutions for coupled van del pol oscillatorys of two-degrees-of-freedom by homotopy analysis method[J].Science&Technology Review,2008,26(22):21–25.(钱有华,张伟.两自由度耦合van del Pol振子周期解的同伦分析方法[J].科技导报,2008,26(22):21–25.)
[11]MOHAMMED I,SELJAK U.Analytic model for the matter power spectrum,its covariance matrix and baryonic effects[J].Monthly Notices of the Royal Astronomical Society,2017,445(4):3382–3400.
[12]JACKEL B.Characterization of auroral radar power spectra and autocorrelation functions[J].Radio Science,2016,35(4):1009–1023.
[13]KONDRASHOV D,SHPRITS Y,GHIL M.Gap filling of solar wind data by singular spectrum analysis[J].Geophysical Research Letters,2015,37(37):78–82.
[14]LU Chenglong,KUANG Cuilin,YI Zhonghai,et al.Singular spectrum analysis filter method for mitigation of GPS multipath error[J].Geomatics and Information Science of Wuhan University,2015,40(7):924–931.(卢辰龙,匡翠林,易重海,等.奇异谱分析滤波法在消除GPS多路径中的应用[J].武汉大学学报(信息科学版),2015,40(7):924–931.)
[15]YUE Shun,LI Xiaoqi,ZHAI Changzhi.Determining the singular spectrum embedding dimension based on an improved Cao algorithm[J].Engineering of Surveying and Mapping,2015,24(3):64–68.(岳顺,李小奇,翟长治.基于改进Cao算法确定奇异谱嵌入维数及应用[J].测绘工程,2015,24(3):64–68.)
[16]YOU Fengchun,DING Yuguo,ZHOU Yu,et al.Application of SVD and SSCA in the prediction of summer precipitation diagnosisin North China[J].Journal of Applied Meteorological Science,2003,14(2):176–187.(尤凤春,丁裕国,周煜,等.奇异值分解和奇异交叉谱分析方法在华北夏季降水诊断中的应用[J].应用气象学,2003,14(2):176–187.)
[17]BENOWITZ B,SHIELD M,DEODATIS G.Determining evolutionary spectra from non-stationary autocorrelation functions[J].Probabilistic Engineering Mechanics,2015,41:73–88.
[18]NI Lili,ZHANG Yanlan.The determination of the best embedding dimension of bridge deformation time series based on G--P method and Cao method[J].Northern Communications,2017,(5):5–7.(泥立丽,张艳兰.基于G--P法和Cao法的桥梁变形时间序列最佳嵌入维数的确定[J].北方交通,2017,(5):5–7.)
[19]GU C,ROHLING J,LIANG X,et al.Impact of dispersed coupling strength on the free running periods of circadian rhythms[J].Physical Review E,2016,93(3-1):032414.
[20]WU Yilin,SHEN Zhiping.The tracking problem in networked systems with periodic signal reference input[J].Control Theory&Applications,2016,33(5):685–693.(邬依林,沈志萍.网络化系统周期信号的跟踪[J].控制理论与应用,2016,33(5):685–693.)
[21]WANG Lei,XU Genqi.Stabilization of the complex wave network with two circuits[J].Control Theory&Applications,2011,28(6):759–762.(王雷,许跟起.双回路复杂杂波网络的镇定[J].控制理论与应用,2011,28(6):759–762.)
1由于已证实第1阶段并不存在均值回归,故未选用第1阶段数据.
2 指欧洲排放权配额.
3 经典谱估计又分为布莱克曼–杜基谱估计器,Blackman和J.Tukey,BT谱估计器)(间接法)和周期图法(直接法);现代谱估计可分为参数模型谱估计和非参数模型谱估计.
4 Tukey根据Wiener-Khintchine定理提出了对有限长数据进行谱估计的自相关法,即利用有限长度数据估计自相关函数,再对该自相关函数求傅立叶变换,从而得到谱的估计.
5月度数据是反映时间序列中长周期的依据.
6特征值分别表示信号不同阶数的主成分,特征值较小的成分反映噪声项.