Current research on carbon markets primarily focuses on carbon emission reduction effects and economic impact assessments, with relatively limited research on carbon market investment. Carbon emissions trading is an important tool for addressing climate change and promoting low-carbon transformation. Investment in the carbon market can promote optimal resource allocation, balance risk and return, and improve investment efficiency. Therefore, research on portfolio optimization in the carbon market is not only a means of risk management, but also an important approach to promote carbon market maturity and achieve sustainable investment goals.
This paper focuses on portfolio optimization in China’s carbon market. First, based on the carbon price data from seven carbon emissions trading pilot regions in China from 2017 to 2021, a possibility mean-semi-absolute deviation portfolio optimization model is constructed considering realistic constraints. The model introduces trapezoidal fuzzy numbers to address asset uncertainty, measuring expected returns by possibility mean and risk by possibility semi-absolute deviation. Second, an improved artificial bee colony algorithm (IABC) is designed to solve the constructed model and enhance solution efficiency. Finally, the empirical study conducts both in-sample analysis and out-of-sample testing. The in-sample analysis explores optimal carbon asset allocation strategies for investors under different expected returns and risk preferences, while out-of-sample testing employs a rolling window method to compare the performance of the proposed model with the classical mean-variance model in terms of Sharpe ratio, downside risk, Sortino ratio, and other metrics, verifying the effectiveness and robustness of the model. The results show significant differences in returns, volatility, skewness, and other factors among China’s seven carbon emissions trading pilot markets. As expected returns increase, investment risk rises synchronously, and the allocation proportion to high volatility carbon market increases. When risk aversion level rises, funds flow more toward carbon markets with stable returns. Investors should dynamically adjust their asset allocation structure according to risk appetite and expected returns. Out-of-sample tests demonstrate that the model and algorithm have significant advantages in reducing portfolio downside risk and enhancing investment robustness.
Compared with previous literature, this paper contributes in the following three aspects: First, through the portfolio optimization model, investors can effectively allocate capital and choose appropriate carbon emissions exchanges for investment. This optimized investment approach can guide more capital into the carbon trading market, supporting financing needs of low-carbon and green projects and providing stronger capital support for low-carbon economic development. Second, carbon markets have different risk and return characteristics due to policy, market fluctuations, and other factors. Through portfolio optimization, investments can be diversified across different carbon emission exchanges to effectively control overall risk. Finally, research on carbon market portfolio optimization can promote financial innovation in the carbon market and encourage the introduction of more carbon financial products, such as carbon index funds, carbon futures and carbon bonds, to meet investors’ diverse needs.
This study provides theoretical support for optimal portfolio research in the carbon market, offers decision-making reference for investors participating in the carbon market, and contributes to the healthy development of China’s carbon market.
YAOD, SHUAIA Q, YANGA J. Volatility characteristics of carbon emission trading markets in Guangdong and Hubei provinces based on a score-driven model[J]. Practice and Understanding of Mathematics,2022,52(12):267-274.
ZHANGX X, SUNS J. Can carbon emission trading policies promote carbon reduction: a study based on prefecture-level panel data[J]. Nankai Economic Studies,2024,40(2):160-178.
[7]
CHENZ H, HEY, LIAON. Can carbon emission trading policy enhance the synergistic emission reduction of carbon dioxide and air pollutants?A comparative study considering different pollutants[J]. Energy,2024,305(20):132364.
[8]
SHANP, ZHANGY, ZHANGL, et al. The impact of carbon trading pilots on carbon emission reduction in China’s power industry: a study using the generalized synthetic control method[J].Energy Reports,2025,13(1):3605-3615.
LIUC M, SUNZ, ZHANGJ. The carbon reduction policy effect of China’s carbon emission trading pilot[J]. China Population, Resources and Environment,2019,29(11):49-58.
TANJ, ZHANGJ H. Has the carbon trading mechanism forced industrial structure upgrading: an analysis based on the synthetic control method[J]. Economic and Management Research,2018,39(12):104-119.
SHAOS, LIX. Can market-oriented low-carbon policies promote high-quality economic development: evidence from carbon emission trading pilots [J]. Guangdong Social Sciences,2022,39(2):33-45.
[19]
BATTENJ A, MADDOXG E, YOUNGM R. Does weather, or energy prices, affect carbon prices?[J]. Energy Economics,2021,96(4):105016.
[20]
LIUZ B, HUANGS. Carbon option price forecasting based on modified fractional Brownian motion optimized by GARCH model in carbon emission trading[J]. The North American Journal of Economics and Finance,2021,55(1):101307.
[21]
ORTIZR, CONTRERASM, MELLADOC. Regression, multicollinearity and Markowitz[J]. Finance Research Letters,2023,58(8):104550.
[22]
HOSSEINI-NODEHZ, KHANJANI-SHIRAZR, PARDALOSP M. Portfolio optimization using robust mean absolute deviation model: Wasserstein metric approach[J]. Finance Research Letters,2023,54(4):103735.
[23]
SEHGALR, JAGADESHP. Data-driven robust portfolio optimization with semi mean absolute deviation via support vector clustering[J]. Expert Systems with Applications,2023,224(14):120000.
[24]
VERCHERE, BERMÚDEZJ D. Portfolio optimization using a credibility mean-absolute semi-deviation model[J]. Expert Systems with Applications,2015,42(20):7121-7131.
[25]
YUB, ZHENGZ J, CAIM J, et al. FRCM: a fuzzy rough c-means clustering method[J].Fuzzy Sets and Systems,2024,480(7):108860.
[26]
KHARRIMM. Multi-period fuzzy portfolio optimization model subject to real constraints[J].EURO Journal on Decision Processes,2023,11(1):100041.
ZHANGP, YES N.Portfolio optimization with stochastic fuzzy mean-semi absolute deviation under mental account theory[J]. Fuzzy Systems and Mathematics,2022,36(3):154-163.
LIUY J, WUJ D, ZHANGW G. A hybrid portfolio optimization model with flexible time horizons[J]. Chinese Journal of Management Science,2024,32(1):23-30.
[31]
JALOTAH, MANDALP K, THAKURM, et al. A novel approach to incorporate investor’s preference in fuzzy multi-objective portfolio selection problem using credibility measure[J]. Expert Systems with Applications,2023,212(2):118583.
[32]
ZHANGY, LIUW L, YANGX Y. An automatic trading system for fuzzy portfolio optimization problem with sell orders[J]. Expert Systems with Applications,2022,187(1):115822.
[33]
GONGX M, MINL Y, YUC R. Multi-period portfolio selection under the coherent fuzzy environment with dynamic risk-tolerance and expected-return levels[J]. Applied Soft Computing,2022,114(1):108104.
[34]
DYMOVAL, KACZMAREKK, SEVASTJANOVP. A new approach to the bi-criteria multi-period fuzzy portfolio selection[J]. Knowledge-Based Systems, 2021,234(24):107582.
[35]
PAHADEJ K, JHA M. A hybrid fuzzy-SCOOT algorithm to optimize possibilistic mean semi-absolute deviation model for optimal portfolio selection[J]. International Journal of Fuzzy Systems,2022,24(4):1958-1973.
[36]
LIX, ZHANGS P, SHAOP. Discrete artificial bee colony algorithm with fixed neighborhood search for traveling salesman problem[J]. Engineering Applications of Artificial Intelligence,2024,131(5):107816.
[37]
SONGX Y, ZHAOM, XINGS Y. A multi-strategy fusion artificial bee colony algorithm with small population[J]. Expert Systems with Applications,2020,142(4):112921.
[38]
DENGY L, XUH F, WUJ. Optimization of blockchain investment portfolio under artificial bee colony algorithm[J]. Journal of Computational and Applied Mathematics,2021,385(5):113199.
SHIJ P, XUY C, GUX, et al. An improved artificial bee colony algorithm for the inverse kinematics problem of redundant manipulators[J]. Journal of Mechanical Engineering,2024,60(15):60-70.
LIZ Y, HUANGT, CHENS M, et al. A survey of constrained optimization evolutionary algorithms[J]. Journal of Software,2017,28(6):1529-1546.
[43]
ARDIAD, GUIDOTTIE, KROENCKET A. Efficient estimation of bid-ask spreads from open, high, low, and close prices[J]. Journal of Financial Economics,2024,161(11):103916.
[44]
GELMINIM, UBERTIP. The equally weighted portfolio still remains a challenging benchmark[J]. International Economics,2024,179(3):100525.