Green cryptocurrencies and artificial intelligence (AI)-coins have emerged as prominent digital asset classes amid growing interest in sustainable finance and technological innovation. Despite their rapid development, the empirical relationship between these two segments remains largely unexplored. To address this gap, assess dynamic risk connectedness, and identify net transmitter and net receiver spillovers between green and AI cryptocurrencies, we first use a time-varying parameter vector autoregressive framework to characterize the dynamic interdependencies across time and frequency. To further account for nonlinear dependence and improve interpretability for portfolio construction, we integrate eXtreme Gradient Boosting (XGBoost) with SHapley Additive exPlanations (SHAP) and implement an AI-driven portfolio-optimization strategy spanning green cryptocurrencies and AI tokens. Empirically, the central findings indicate that Ethereum (ETH), the dominant green cryptocurrency, persistently serves as a net transmitter of volatility to other green assets and to the five major AI-crypto markets. SHAP-based attribution further shows that the portfolio-contribution metric (Zp), together with Ripple (XRP) and ETH, ranks among the most influential features. Overall, the proposed XGBoost-based approach captures nonlinear interactions among momentum, liquidity, and volatility, enhancing forecasting and portfolio allocation. When evaluated out-of-sample over 438 daily-rebalanced trading days (August 13, 2024–October 24, 2025), the machine learning methodology attains an annualized Sharpe ratio of approximately 2.08 at mean daily volatility near 1.04%. It also achieves the highest terminal wealth among the benchmark strategies (USD 17,812 from a USD 10,000 baseline), yielding a 78.1% cumulative return and superior risk-adjusted performance relative to competing strategies. These findings provide economically intuitive evidence of cross-segment spillovers and yield policy-relevant implications for machine-learning-based portfolio allocation across sustainable and AI cryptocurrency markets. They also offer useful insights for regulators with respect to systemic risk surveillance, contagion monitoring, and the preservation of market integrity.
Citation: Po-Sheng Ko, Kuo-Shing Chen. Unveiling dynamic network connectedness and machine-intelligent portfolio optimization in green cryptocurrency and AI-token markets[J]. Networks and Heterogeneous Media, 2026, 21(2): 632-668. doi: 10.3934/nhm.2026028
Green cryptocurrencies and artificial intelligence (AI)-coins have emerged as prominent digital asset classes amid growing interest in sustainable finance and technological innovation. Despite their rapid development, the empirical relationship between these two segments remains largely unexplored. To address this gap, assess dynamic risk connectedness, and identify net transmitter and net receiver spillovers between green and AI cryptocurrencies, we first use a time-varying parameter vector autoregressive framework to characterize the dynamic interdependencies across time and frequency. To further account for nonlinear dependence and improve interpretability for portfolio construction, we integrate eXtreme Gradient Boosting (XGBoost) with SHapley Additive exPlanations (SHAP) and implement an AI-driven portfolio-optimization strategy spanning green cryptocurrencies and AI tokens. Empirically, the central findings indicate that Ethereum (ETH), the dominant green cryptocurrency, persistently serves as a net transmitter of volatility to other green assets and to the five major AI-crypto markets. SHAP-based attribution further shows that the portfolio-contribution metric (Zp), together with Ripple (XRP) and ETH, ranks among the most influential features. Overall, the proposed XGBoost-based approach captures nonlinear interactions among momentum, liquidity, and volatility, enhancing forecasting and portfolio allocation. When evaluated out-of-sample over 438 daily-rebalanced trading days (August 13, 2024–October 24, 2025), the machine learning methodology attains an annualized Sharpe ratio of approximately 2.08 at mean daily volatility near 1.04%. It also achieves the highest terminal wealth among the benchmark strategies (USD 17,812 from a USD 10,000 baseline), yielding a 78.1% cumulative return and superior risk-adjusted performance relative to competing strategies. These findings provide economically intuitive evidence of cross-segment spillovers and yield policy-relevant implications for machine-learning-based portfolio allocation across sustainable and AI cryptocurrency markets. They also offer useful insights for regulators with respect to systemic risk surveillance, contagion monitoring, and the preservation of market integrity.
| [1] |
D. B. Olawade, O. Z. Wada, A. C. David-Olawade, O. Fapohunda, A. O. Ige, J. Ling, Artificial intelligence potential for net zero sustainability: Current evidence and prospects, Next Sustainability, 4 (2024), 100041. https://doi.org/10.1016/j.nxsust.2024.100041 doi: 10.1016/j.nxsust.2024.100041
|
| [2] |
T. Lim, Environmental, social, and governance (ESG) and artificial intelligence in finance: State-of-the-art and research takeaways, Artif. Intell. Rev., 57 (2024), 76. https://doi.org/10.1007/s10462-024-10708-3 doi: 10.1007/s10462-024-10708-3
|
| [3] |
R. Patel, S. Kumar, E. Bouri, N. Iqbal, Spillovers between green and dirty cryptocurrencies and socially responsible investments around the war in Ukraine, Int. Rev. Econ. Finance, 87 (2023), 143–162. https://doi.org/10.1016/j.iref.2023.04.013 doi: 10.1016/j.iref.2023.04.013
|
| [4] |
M. Zeng, W. Zhang, Green finance: The catalyst for artificial intelligence and energy efficiency in Chinese urban sustainable development, Energy Econ., 139 (2024), 107883. https://doi.org/10.1016/j.eneco.2024.107883 doi: 10.1016/j.eneco.2024.107883
|
| [5] |
L. Pham, T. L. D. Huynh, W. Hanif, Time-varying asymmetric spillovers among cryptocurrency, green and fossil-fuel investments, Glob. Finance J., 58 (2023), 100891. https://doi.org/10.1016/j.gfj.2023.100891 doi: 10.1016/j.gfj.2023.100891
|
| [6] |
B. Bedowska-Sojka, A. Kliber, Do investors in dirty and clean cryptocurrencies care about energy efficiency in the same way? Finance Res. Lett., 67 (2024), 105852. https://doi.org/10.1016/j.frl.2024.105852 doi: 10.1016/j.frl.2024.105852
|
| [7] |
A. Kliber, B. Bedowska-Sojka, Proof-of-work versus proof-of-stake coins as possible hedges against green and dirty energy, Energy Econ., 138 (2024), 107820. https://doi.org/10.1016/j.eneco.2023.107820 doi: 10.1016/j.eneco.2023.107820
|
| [8] | A. Yakovenko, Solana: A new architecture for a high performance blockchain v0.8.13, White paper, (2018). Available from: http://tinyurl.com/hmks9mx7. |
| [9] | S. Alizadeh, M. Khabbazian, Solana's transaction network: Analysis, insights, and comparison, EPJ Data Sci., 14 (2025), 1–16. |
| [10] |
T. L. D. Huynh, E. Hille, M. A. Nasir, Diversification in the age of the 4th industrial revolution: The role of artificial intelligence, green bonds and cryptocurrencies, Technol. Forecast. Soc. Change, 159 (2020), 120188. https://doi.org/10.1016/j.techfore.2020.120188 doi: 10.1016/j.techfore.2020.120188
|
| [11] |
S. Demiralay, H. G. Gencer, S. Bayraci, How do artificial intelligence and robotics stocks co-move with traditional and alternative assets in the age of the 4th industrial revolution? Implications and insights for the COVID-19 period, Technol. Forecast. Soc. Change, 171 (2021), 120989. https://doi.org/10.1016/j.techfore.2021.120989 doi: 10.1016/j.techfore.2021.120989
|
| [12] |
F. Jareno, I. Yousaf, Artificial intelligence-based tokens: Fresh evidence of connectedness with artificial intelligence-based equities, Int. Rev. Financ. Anal., 89 (2023), 102826. https://doi.org/10.1016/j.irfa.2023.102826 doi: 10.1016/j.irfa.2023.102826
|
| [13] |
M. Abdullah, P. K. Sarker, E. J. A. Abakah, A. K. Tiwari, M. Z. Rehman, Tail risk intersection between tech-tokens and tech-stocks, Glob. Financ. J., 61 (2024), 100989. https://doi.org/10.1016/j.gfj.2023.100989 doi: 10.1016/j.gfj.2023.100989
|
| [14] |
A. Husain, K. J. Yii, C. C. Lee, Are green cryptocurrencies really green? New evidence from wavelet analysis, J. Clean. Prod., 417 (2023), 137985. https://doi.org/10.1016/j.jclepro.2023.137985 doi: 10.1016/j.jclepro.2023.137985
|
| [15] |
K. Duan, Y. Zhao, A. Urquhart, Y. Huang, Do clean and dirty cryptocurrencies connect with financial assets differently? The role of economic policy uncertainty, Energy Econ., 127 (2023), 107079. https://doi.org/10.1016/j.eneco.2023.107079 doi: 10.1016/j.eneco.2023.107079
|
| [16] |
M. A. Naeem, T. T. H. Nguyen, S. Karim, B. M. Lucey, Extreme downside risk transmission between green cryptocurrencies and energy markets: The diversification benefits, Finance Res. Lett., 58 (2023), 104263. https://doi.org/10.1016/j.frl.2022.104263 doi: 10.1016/j.frl.2022.104263
|
| [17] |
F. Ali, M. U. Khurram, A. Sensoy, X. V. Vo, Green cryptocurrencies and portfolio diversification in the era of greener paths, Renewable Sustainable Energy Rev., 191 (2024), 114137. https://doi.org/10.1016/j.rser.2023.114137 doi: 10.1016/j.rser.2023.114137
|
| [18] |
J. Almeida, T. C. Gonçalves, The AI revolution: Are crypto markets more efficient after ChatGPT 3? Finance Res. Lett., 66 (2024), 105608. https://doi.org/10.1016/j.frl.2024.105608 doi: 10.1016/j.frl.2024.105608
|
| [19] |
A. Saggu, L. Ante, The influence of ChatGPT on artificial intelligence related crypto assets: Evidence from a synthetic control analysis, Finance Res. Lett., 55 (2023), 103993. https://doi.org/10.1016/j.frl.2023.103993 doi: 10.1016/j.frl.2023.103993
|
| [20] | S. Chipolina, Bitcoin leaves the rest of the crypto market behind, FT Cryptofinance Newsletter, (2023). Available from: https://www.ft.com/content/5160b007-5e68-423d-8a6c-a0fdef6199f4?syn-25a6b1a6 = 1. |
| [21] |
L. Ante, E. Demir, The ChatGPT effect on AI-themed cryptocurrencies, Econ. Bus. Lett., 13 (2024), 29–38. https://doi.org/10.17811/ebl.14.1.2024.29-38 doi: 10.17811/ebl.14.1.2024.29-38
|
| [22] |
C. Stoll, L. Klaassen, U. Gallersdorfer, The carbon footprint of Bitcoin, Joule, 3 (2019), 1647–1661. https://doi.org/10.1016/j.joule.2019.05.012 doi: 10.1016/j.joule.2019.05.012
|
| [23] |
B. Ren, B. Lucey, A clean, green haven?—Examining the relationship between clean energy, clean and dirty cryptocurrencies, Energy Econ., 109 (2022), 105951. https://doi.org/10.1016/j.eneco.2022.105951 doi: 10.1016/j.eneco.2022.105951
|
| [24] |
L. Pham, S. Karim, M. A. Naeem, C. Long, A tale of two tails among carbon prices, green and non-green cryptocurrencies, Int. Rev. Financ. Anal., 82 (2022), 102139. https://doi.org/10.1016/j.irfa.2022.102139 doi: 10.1016/j.irfa.2022.102139
|
| [25] |
K. S. Chen, W. C. Ong, Dynamic correlations between Bitcoin, carbon emission, oil and gold markets: New implications for portfolio management, AIMS Math., 9 (2024), 1403–1433. https://doi.org/10.3934/math.2024069 doi: 10.3934/math.2024069
|
| [26] |
S. Long, H. Tian, Z. Li, Dynamic spillovers between uncertainties and green bond markets in the US, Europe, and China: Evidence from the quantile VAR framework, Int. Rev. Financ. Anal., 84 (2022), 102416. https://doi.org/10.1016/j.irfa.2022.102416 doi: 10.1016/j.irfa.2022.102416
|
| [27] |
K.C. Lu, K.S. Chen, Uncovering information linkages between Bitcoin, sustainable finance and the impact of COVID-19: Fractal and entropy analysis, Fractal Fract., 7 (2023), 424. https://doi.org/10.3390/fractalfract7060424 doi: 10.3390/fractalfract7060424
|
| [28] |
J. Wu, J. P. Li, C. W. Su, Can green bond hedges climate policy uncertainty in the United States? Econ. Anal. Policy, 82 (2024), 1158–1176. https://doi.org/10.1016/j.eap.2023.08.015 doi: 10.1016/j.eap.2023.08.015
|
| [29] |
I. Yousaf, O. M. Ohikhuare, Y. Li, Y. Li, Interconnectedness between electricity and artificial intelligence-based markets during the crisis periods: Evidence from the TVP-VAR approach, Energy Econ., 139 (2024), 107885. https://doi.org/10.1016/j.eneco.2024.107885 doi: 10.1016/j.eneco.2024.107885
|
| [30] |
S. B. Jabeur, G. Gozgor, H. Rezgui, K. S. Mohammed, Dynamic dependence between quantum computing stocks and Bitcoin: Portfolio strategies for a new era of asset classes, Int. Rev. Financ. Anal., 95 (2024), 103478. https://doi.org/10.1016/j.irfa.2023.103478 doi: 10.1016/j.irfa.2023.103478
|
| [31] |
C. C. Wu, W. P. Chen, What's an AI name worth? The impact of AI ETFs on their underlying stocks, Finance Res. Lett., 46 (2022), 102474. https://doi.org/10.1016/j.frl.2021.102474 doi: 10.1016/j.frl.2021.102474
|
| [32] |
A. K. Tiwari, E. J. A. Abakah, T. L. Le, D. I. Leyva-de la Hiz, Markov-switching dependence between artificial intelligence and carbon price: The role of policy uncertainty in the era of the 4th industrial revolution and the effect of COVID-19 pandemic, Technol. Forecast. Soc. Change, 163 (2021), 120434. https://doi.org/10.1016/j.techfore.2020.120434 doi: 10.1016/j.techfore.2020.120434
|
| [33] |
S. Sharma, A. K. Tiwari, S. Nasreen, Are FinTech, robotics, and blockchain index funds providing diversification opportunities with emerging markets? Lessons from pre and post outbreak of COVID-19, Electron. Commer. Res., 24 (2024), 341–370. https://doi.org/10.1007/s10660-022-09611-2 doi: 10.1007/s10660-022-09611-2
|
| [34] |
E. Jacquier, A. Kane, A. J. Marcus, Geometric or Arithmetic Mean: A Reconsideration, Financ. Anal. J., 59 (2003), 46–53. https://doi.org/10.2469/faj.v59.n6.2574 doi: 10.2469/faj.v59.n6.2574
|
| [35] |
A. Moreira, T. Muir, Volatility-managed portfolios, J. Finance, 72 (2017), 1611–1644. https://doi.org/10.1111/jofi.12513 doi: 10.1111/jofi.12513
|
| [36] | C. R. Harvey, E. Hoyle, R. Korgaonkar, S. Rattray, M. Sargaison, O. Van Hemert, The impact of volatility targeting, SSRN Electron. J., (2018), 3175538. |
| [37] |
K. Daniel, T. J. Moskowitz, Momentum crashes, J. Financ. Econ., 122 (2016), 221–247. https://doi.org/10.1016/j.jfineco.2015.12.002 doi: 10.1016/j.jfineco.2015.12.002
|
| [38] |
S. Gu, B. Kelly, D. Xiu, Empirical asset pricing via machine learning, Rev. Financ. Stud., 33 (2020), 2223–2273.https://doi.org/10.1093/rfs/hhaa009 doi: 10.1093/rfs/hhaa009
|
| [39] |
L. Chen, M. Pelger, J. Zhu, Deep learning in asset pricing, Manage. Sci., 70 (2024), 714–750. https://doi.org/10.1287/mnsc.2023.4695 doi: 10.1287/mnsc.2023.4695
|
| [40] |
J. Jang, N. Seong, Deep reinforcement learning for stock portfolio optimization by connecting with modern portfolio theory, Expert Syst. Appl., 218 (2023), 119556. https://doi.org/10.1016/j.eswa.2023.119556 doi: 10.1016/j.eswa.2023.119556
|
| [41] |
F. X. Diebold, K. Yilmaz, Better to give than to receive: Predictive directional measurement of volatility spillovers, Int. J. Forecast., 28 (2012), 57–66. https://doi.org/10.1016/j.ijforecast.2011.02.006 doi: 10.1016/j.ijforecast.2011.02.006
|
| [42] |
H. H. Pesaran, Y. Shin, Generalized impulse response analysis in linear multivariate models, Econ. Lett., 58 (1998), 17–29. https://doi.org/10.1016/S0165-1765(97)00214-0 doi: 10.1016/S0165-1765(97)00214-0
|
| [43] |
N. Antonakakis, I. Chatziantoniou, D. Gabauer, Refined measures of dynamic connectedness based on time-varying parameter vector autoregressions, J. Risk Financ. Manage., 13 (2020), 84. https://doi.org/10.3390/jrfm13040084 doi: 10.3390/jrfm13040084
|
| [44] |
Q. Ji, E. Bouri, C. K. M. Lau, D. Roubaud, Dynamic connectedness and integration in cryptocurrency markets, Int. Rev. Financ. Anal., 63 (2019), 257–272. https://doi.org/10.1016/j.irfa.2018.12.002 doi: 10.1016/j.irfa.2018.12.002
|
| [45] |
I. O. Fasanya, O. Oyewole, T. Odudu, Returns and volatility spillovers among cryptocurrency portfolios, Int. J. Manag. Financ., 17 (2021), 327–341. https://doi.org/10.1108/IJMF-02-2019-0074 doi: 10.1108/IJMF-02-2019-0074
|
| [46] |
G. Koop, M. H. Pesaran, S. M. Potter, Impulse response analysis in nonlinear multivariate models, J. Econom., 74 (1996), 119–147. https://doi.org/10.1016/0304-4076(95)01753-4 doi: 10.1016/0304-4076(95)01753-4
|
| [47] |
K. Phoon, F. Koh, Robo-Advisors and wealth management, J. Altern. Invest., 20 (2017), 79–94. https://doi.org/10.3905/jai.2018.20.3.079 doi: 10.3905/jai.2018.20.3.079
|
| [48] | T. Chen, C. Guestrin, XGBoost: A scalable tree boosting system, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, (2016), 785–794. https://doi.org/10.1145/2939672.2939785 |
| [49] | S. M. Lundberg, S. Lee, A unified approach to interpreting model predictions, in Advances in Neural Information Processing Systems, 30 (2017), 4765–4774. |
| [50] |
S. M. Lundberg, G. Erion, H. Chen, A. DeGrave, J. M. Prutkin, B. Nair, et al., From local explanations to global understanding with explainable AI for trees, Nat. Mach. Intell., 2 (2020), 56–67. https://doi.org/10.1038/s42256-019-0138-9 doi: 10.1038/s42256-019-0138-9
|
| [51] |
R. Fry-McKibbin, C. Y. L. Hsiao, Extremal dependence tests for contagion, Econ. Rev., 37 (2018), 626–649. https://doi.org/10.1080/07474938.2018.1452356 doi: 10.1080/07474938.2018.1452356
|
| [52] |
K. J. Forbes, R. Rigobon, No contagion, only interdependence: Measuring stock market co-movements, J. Finance, 57 (2002), 2223–2261. https://doi.org/10.1111/0022-1082.00494 doi: 10.1111/0022-1082.00494
|
| [53] |
R. Fry, V. L. Martin, C. Tang, A new class of tests of contagion with applications, J. Bus. Econ. Stat., 28 (2010), 423–437. https://doi.org/10.1198/jbes.2009.08098 doi: 10.1198/jbes.2009.08098
|
| [54] |
P. K. Narayan, Understanding exchange rate shocks during COVID-19, Finance Res. Lett., 45 (2022), 102181.https://doi.org/10.1016/j.frl.2021.102181 doi: 10.1016/j.frl.2021.102181
|
| [55] |
S. Khademalomoom, P. K. Narayan, Intraday effects of the currency market, J. Int. Financ. Mark. Inst. Money, 58 (2019), 65–77. https://doi.org/10.1016/j.intfin.2018.09.008 doi: 10.1016/j.intfin.2018.09.008
|
| [56] |
J. Barunik, T. Krehlik, Measuring the frequency dynamics of financial connectedness and systemic risk, J. Financ. Econom., 16 (2018), 271–296. https://doi.org/10.1093/jjfinec/nby001 doi: 10.1093/jjfinec/nby001
|
| [57] |
H. Markowitz, Portfolio selection, J. Finance, 7 (1952), 77–91. https://doi.org/10.2307/2975974 doi: 10.2307/2975974
|
| [58] |
H. T. Wu, M. L. Yueh, Cryptocurrency risk management using Levy processes and time-varying volatility, Rev. Quant. Financ. Account., 66 (2026), 33–61. https://doi.org/10.1007/s11156-025-01393-6 doi: 10.1007/s11156-025-01393-6
|
| [59] | A. Phiri, I. Anyikwa, A multiscale analysis of returns and volatility spillovers in cryptocurrency markets: A post-COVID perspective, Invest. Anal. J., 54 (2025), 21–41. https://hdl.handle.net/10520/ejc-invest_v54_n1_a21 |
| [60] |
T. J. Moskowitz, Y. H. Ooi, L. H. Pedersen, Time series momentum, J. Financ. Econ., 104 (2012), 228–250. https://doi.org/10.1016/j.jfineco.2011.11.003 doi: 10.1016/j.jfineco.2011.11.003
|
Figures(18) / Tables(5)
Po-Sheng Ko, Kuo-Shing Chen. Unveiling dynamic network connectedness and machine-intelligent portfolio optimization in green cryptocurrency and AI-token markets[J]. Networks and Heterogeneous Media, 2026, 21(2): 632-668. doi: 10.3934/nhm.2026028
DownLoad: