在减少温室气体排放方面，中国正在面临的国际和国内压力与日俱增。2020年9月22日，国家主席习近平提出，中国力争在2030年前实现碳达峰，在2060年前实现碳中和。为了进一步促进二氧化碳的减排，碳排放权交易作为一种新的市场机制应运而生，许多国家和地区都参与其中。我国陆续建立了八个碳试点市场，全国碳排放权交易市场于2021年7月16日启动上线。但是我国碳排放权价格形成机制尚未完全形成，因此需要对碳排放权价格的各个影响因素进行灵敏度分析，以便于充分掌握市场价格形成机制。

碳排放权价格是碳市场运行状况的直接表现，经济、环境、能源、政策等因素都会影响碳排放权的价格，同时随着碳金融衍生品的推出，更需要着重关注整体政治、经济、社会等环境因素对碳排放权价格的影响。本文选取我国八个有代表性的碳试点市场，从传统能源价格、新能源价格、宏观经济、天气因素、传统金融市场、国际碳资产价格六个方面，对14个影响因素进行研究，建立高斯过程回归模型，最后采用Sobol灵敏度分析方法，对预测输出的估计方差进行定量分析，精确化分析不同影响因素的作用大小，结合碳试点市场的经济环境提出相应问题，以促进我国区域间协调减排、稳定碳价波动、贯彻碳中和政策，为完善全国统一碳市场提供参考。

研究结果表明，14个影响因素的灵敏度指数存在差异，但部分碳试点市场之间存在相似性。总体来看，欧盟碳排放配额、多晶硅、布伦特原油、沪深300指数是灵敏度指数最高的四个影响因素；对比各碳试点市场后发现，不同市场的影响因素不同，而同一影响因素在不同碳试点市场发挥的作用也不同。同时，各影响因素的灵敏度指数也不高，表明碳价对各影响因素的反应程度较低，反应机制不足；除此之外，结合碳试点市场的经济环境，发现我国传统能源行业转型过慢且目前对传统能源依赖程度较高、传统金融市场与碳市场联动性较弱、天气因素与低碳产业的连接程度不高、不同地区新能源行业发展水平各异等问题。提出了保障传统能源行业逐步转型、加强传统金融市场与碳市场的联动性、开展气候投融资地方试点、促进各地区新能源行业平衡稳定发展的对策建议。

China is facing increasing international and domestic pressure to reduce its greenhouse gas emissions. On September 22, 2020, President Xi proposed at the 75th United Nations General Assembly that China's carbon dioxide emissions should reach the peak by 2030 and strive to achieve the goal of carbon neutral by 2060. Carbon emission trading is a market mechanism launched by countries or regions in order to promote the reduction of carbon dioxide. China has set up eight carbon pilot markets, and a national carbon emission trading market launched in 16 July. However, the price formation mechanism of carbon price has not been fully formed in China, so it is necessary to conduct sensitivity analysis on each influencing factor of carbon price in order to fully grasp the market price formation mechanism.

The carbon price is a direct manifestation of the operating status of the carbon market. A series of factors such as energy, social environment, and national policies will have varying degrees of impact on the transaction price of carbon dioxide transactions. Especially with the future carbon the launch of financial derivatives, more need to focus on the overall political, economic, social and other environmental factors on the influence of the carbon market prices. This paper selects eight representative carbon markets in China, including traditional energy prices, new energy prices, macroeconomics, weather factors, traditional financial markets, and international carbon futures prices, as well as 14 influencing factors for research, using Gaussian process regression Model modeling.

Finally, the Sobol sensitivity analysis method is used to quantitatively analyze the estimated variance of the forecast output, accurately analyze the effect of different influencing factors, and conduct a horizontal comparison of the sensitivity of the factors affecting the price of carbon emission rights in China to study different regional markets. Questions are raised to promote coordinated emission reduction among regions in China, stabilize carbon price fluctuations, implement carbon neutrality, and promote the smooth operation of the national 

unified carbon market, so as to promote coordinated emission reduction among regions in China according to the actual development of each carbon market combined with the external economic environment in which different carbon pilot markets are located, and provide reference for the improvement of the national unified carbon market.

The results of sensitivity analysis show that the first order sensitivity indices of the 14 influencing factors vary in size, but some trends are similar among some carbon pilot markets. In general, EUA, polysilicon price, Brent crude oil, HS300 index and air quality index are the five factors with the highest sensitivity index. Through the horizontal comparison of the eight carbon pilot markets, it is found that the development degree of different carbon pilot markets in China is different, and there is no coordination between regions. The influencing factors of carbon price are also different, and the same influencing factor plays different roles in different carbon pilot markets. At the same time, the sensitivity index of the sensitive factors is not high, indicating that the reaction degree of carbon price to each influencing factor is low, and the reaction mechanism is insufficient.

In addition, combined with the external economic environment in which different carbon pilot markets are located, it is found that the transformation of my country's traditional energy industry is too slow, the current dependence on traditional energy is relatively high, the linkage between traditional financial markets and carbon markets is weak, the degree of connection between low-carbon industries and weather factors is not high, the impact of new energy prices on different carbon pilot markets is quite different, and the development level of new energy industries in different regions is different. Countermeasures and suggestions are put forward to ensure the gradual transformation of the traditional energy industry, strengthen the linkage between the traditional financial market and the carbon market, carry out local pilot projects for climate investment and financing, and promote the balanced and stable development of the new energy industry in various regions.

[1]林伯强.中国低碳转型[M].北京:科学出版社,2011:35.

[2]欧阳志远,史作廷,石敏俊,杨德伟,龙如银,周宏春,林思佳,郭瑞芳,王宇杰.“碳达峰碳中和”：挑战与对策[J].河北经贸大学学报,2021,42(05):1-11.

[3]周朝波,覃云.碳排放交易试点政策促进了中国低碳经济转型吗?——基于双重差分模型的实证研究[J].软科学,2020,34(10):36-42+55.

[4]Aatola, P., Ollikainen, M., Toppinen, A. Price determination in the EU ETS market:theory and econometric analysis with market fundamentals[J].Energy Econ.,2013, 36: 380–395.

[5]Jonathan A. Batten, Grace E. Maddox, Martin R. Young. Does weather, or energy prices, affect carbon prices? [J]. Energy Economics.,2020, 11:105-116.

[6]Lutz, B.J., Pigorsch, U., Rotfuß, W., Nonlinearity in cap-and-trade systems: the EUA price and its fundamentals [J]. Energy Econ.,2013, 40: 222-232.

[7]Sandra Schusser, Jūratė Jaraitė. Explaining the interplay of three markets: Green certificates, carbon emissions and electricity [J]. Energy Economics.,2018, 71: 1-13.

[8]Deborah Cotton, Lurion De Mello. Econometric analysis of Australian emissions markets and electricity prices [J]. Energy Policy.,2014(74): 475-485.

[9]Xianbing Liu, Zhen Jin. An analysis of the interactions between electricity, fossil fuel and carbon market prices in Guangdong, China. Energy for Sustainable Development.,2020, 55: 82-94.

[10]Mansanet-Bataller M, Pardo A, Valor E. CO₂ prices, energy and weather [J]. The Energy Journal,,2007: 73-92.

[11]Alberola, E., Chevallier, J., Chèze, B., 2008. Price drivers and structural breaks in European carbon prices [J]. Energy Policy, 2008, 36 (2): 787–797.

[12]Benz, E., Trück, S. Modeling the price dynamics of CO2 emission allowances [J]. Energy Econ.,2009, 31 (1): 4–15.

[13]Kaile Zhou, Yiwen Li. Influencing factors and fluctuation characteristics of China’s carbon emission trading price [J]. Physica A: Statistical Mechanics and its Applications, 2019, 15 (524): 459-474.

[14]Meng Han, Lili Ding, Xin Zhao, Wanglin Kang,Forecasting carbon prices in the Shenzhen market, China: The role of mixed-frequency factors[J].Energy,2019, 171:69-76.

[15]Chang-Jing Ji, Yu-Jie Hu, Bao-Jun Tang, Shen Qu. Price drivers in the carbon emissions trading scheme: Evidence from Chinese emissions trading scheme pilots [J]. Journal of Cleaner Production, 2021, 278: 134-169.

[16]Yan Hao, Chengshi Tian. A hybrid framework for carbon trading price forecasting: The role of multiple influence factor [J].Journal of Cleaner Production, 2020, 262: 121-158.

[17]Qi Shaozhou, Zhou Chao-bo, Li Kai, Tang Si-yan. The impact of a carbon trading pilot policy on the low-carbon international competitiveness of industry in China: An empirical analysis based on a DDD model [J]. Journal of Cleaner Production, 2021,281:125-141.

[18]Qi Shaozhou, Cheng Shihan, Cui Jingbo. Environmental and economic effects of China’s carbon market pilots: Empirical evidence based on a DID model [J]. Journal of Cleaner Production,,2021, 279: 172-190.

[19]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,2010,22(06):34-41.

[20]李朝鹏,易兰,杨历.化石能源市场对深圳碳市场价格的影响研究[J].管理现代化,2017,37(05):76-78.

[21]陆敏,苍玉权.影响中国碳价的能源价格因素灰关联研究[J].中国环境管理,2018,10(04):88-92.

[22]王倩,路京京.中国碳配额价格影响因素的区域性差异[J].浙江学刊,2015(04):162-168.

[23]陈欣,刘明,刘延.碳交易价格的驱动因素与结构性断点——基于中国七个碳交易试点的实证研究[J].经济问题,2016(11):29-35.

[24]冯佑帅.空气质量指数与碳排放权价格的交互相关分析[J].宜宾学院学报,2019,19(06):99-105+111.

[25]吕靖烨,杨华,郭泽.基于GA-RS的中国碳排放权价格影响因素的分解研究[J].生态经济,2019,35(11):42-47+130.

[26]李菲菲,钱魏冬,许正松.7个试点省市碳价的影响因素与结构性断点分析[J].西昌学院学报(自然科学版),2020,34(01):27-32.

[27]赵长红,张明明,翟迪.基于试点碳市场的中国碳价驱动因素实证研究[J].经济论坛,2018(12):136-141.

[28]王倩,路京京.人民币汇率对中国碳价的冲击效应——基于区域差异的视角[J].武汉大学学报(哲学社会科学版),2018,71(02):157-165.

[29]杜子平,刘富存.我国区域碳排放权价格及其影响因素研究——基于GA-BP-MIV模型的实证分析[J].价格理论与实践,2018(06):42-45.

[30]吕靖烨,杨华,郭泽.人民币汇率对中国各区域碳价冲击效应的研究——基于VAR模型[J].价格月刊,2019(06):77-84.

[31]易兰,杨历,李朝鹏,任凤涛.欧盟碳价影响因素研究及其对中国的启示[J].中国人口•资源与环境,2017,27(06):42-48.

[32]王巍,韩丹丹.我国碳交易价格影响因素识别及作用机理分析[J].中国经贸导刊(中),2018(26):69-72.

[33]吕靖烨,张超.低碳背景下我国碳排放权市场价格的影响因素分析[J].煤炭经济研究,2019,39(05):31-37.

[34]朱帮助,王平,魏一鸣.基于EMD的碳市场价格影响因素多尺度分析[J].经济学动态,2012(06):92-97.

[35]吴慧娟,张智光.中国碳交易价格低迷的成因:理论模型与实证分析[J/OL].管理现代化,2020(06):75-81.

[36]齐绍洲,王薇.欧盟碳排放权交易体系第三阶段改革对碳价格的影响[J].环境经济研究,2020,5(01):1-20+2.

[37]Bangzhu Zhu, Liqing Huang, Lili Yuan, Shunxin Ye, Ping Wang. Exploring the risk spillover effects between carbon market and electricity market: A bidimensional empirical mode decomposition based conditional value at risk approach [J]. International Review of Economics & Finance, 2020, 67: 163-175.

[38]Ying Ma, Lihua Wang, Tao Zhang. Research on the dynamic linkage among the carbon emission trading, energy and capital markets [J]. Journal of Cleaner Production, 2020, 272: 271-287.

[39]Yazhi Song, Tiansen Liu, Bin Ye, Yin Li. Linking carbon market and electricity market for promoting the grid parity of photovoltaic electricity in China [J]. Energy,2020, 211: 189-202.

[40]Jingye Lyu, Ming Cao, Kuang Wu, Haifeng Li, Ghulam Mohi-ud-din. Price volatility in the carbon market in China [J].Journal of Cleaner Production, 2020, 255: 171-187.

[41]Bangzhu Zhu, Shunxin Ye, Dong Han, Ping Wang, Kaijian He, Yi-Ming Wei, Rui Xie, A multiscale analysis for carbon price drivers[J].Energy Economics, 2019,78:202-216.

[42]Shihong Zeng, Xin Nan, Chao Liu, Jiuying Chen. The response of the Beijing carbon emissions allowance price (BJC) to macroeconomic and energy price indices [J].Energy Policy.,2017, 106: 111-121.

[43]邓海峰,尹瑞龙.碳中和愿景下我国碳排放权交易的功能与制度构造研究[J].北方法学,2022,16(02):5-15.

[44]孟昕,梁志浩.低碳消费偏好下碳排放配额分配方式的减排效应[J].财经问题研究,2022(02):42-51.

[45]宋德勇,朱文博,王班班.中国碳交易试点覆盖企业的微观实证:碳排放权交易、配额分配方法与企业绿色创新[J].中国人口•资源与环境, 2021, 31 (01):37-47.

[46]张跃军,刘景月.我国碳交易制度巩固脱贫攻坚成果的问题及对策[J].中国环境管理,2021,13(01):35-39.

[47]夏睿瞳,宋效军.前瞻性应对碳金融发展挑战[J].中国金融, 2021 (20):73-75.

[48]张叶东.“双碳”目标背景下碳金融制度建设：现状、问题与建议[J].南方金融,2021(11):65-74.

[49]翁智雄.中国实现碳中和远景目标的市场化减排机制研究[J].环境保护, 2021,49(Z1):66-69.

[50]高文文,张占录,张远索.外部性理论下的国土空间规划价值探讨[J].当代经济管理,2021,43(05):80-85.

[51]曹越,彭可人.会计学对科斯定理的完善与推进[J].会计研究,2019(11):34-40.

[52]王露阳,魏庆坡.碳交易市场对接合作初探及对中国的启示[J].生态经济,2016,32(12):53-57.

[53]张晓玲.可持续发展理论:概念演变、维度与展望[J].中国科学院院刊,2018,33(01):10-19.

[54]黄世忠.支撑ESG的三大理论支柱[J].财会月刊,2021(19):3-10.

[55]张锐.欧盟碳市场的运营绩效与经验提炼[J].金融发展研究,2021(10):36-41.

[56]张晶杰,王志轩,雷雨蔚.欧盟碳市场经验对中国碳市场建设的启示[J].价格理论与实践,2020(01):32-36+170.

[57]吕靖烨,范欣雅,吴浩楠.中国碳排放权价格影响因素的参数灵敏度分析[J].软科学,2021,35(05):123-130.

[58]Liu H B,Yang C,Huang M Z,Wang D S, Modeling of subway indoor air quality using Gaussian process regression[J].Journal of Hazardous Materials,2018,359: 266-273.

[59]Sobol,I M.Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates[J].Math. Comput. Simul,2001, 55(1-3):271-280.

[60]Luo J N,Lu W X.Sobol' sensitivity analysis of NAPL-contaminated aquifer remediation process based on multiple surrogates[J].Computers and Geosciences, 2014,67:110-116.

[61]李美水,杨晓华.基于Sobol方法的SWMM模型参数全局敏感性分析[J].中国给水排水,2020,36(17):95-102.

[62]王娟,马义中.考虑相关输入变量的Sobol'指数计算方法和应用[J].系统工程理论与实践,2022,42(03):778-788.

[63]刘松,佘敦先,张利平,丁凯熙,郭梦瑶,陈森林.基于Morris和Sobol的水文模型参数敏感性分析[J].长江流域资源与环境,2019,28(06):1296-1303.