Chong Li

Mr. Chong Li

Xi'an University of Science and Technology, Xi'an, China
orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 December 2024 in Mathematics
Verified
Journal Article
Open access
PDF file
JLJingye Lyu
Jingye Lyu
CLChong Li
Chong Li
WZWenwen Zhou
Wenwen Zhou
JZJinsuo Zhang
Jinsuo Zhang
A scientific assessment of the sensitivity of the Chinese coal market has become an important research topic. This paper combines Gaussian Process Regression (GPR) and Sobol sensitivity analysis to construct a GPR–Sobol hybrid model innovatively applied to the Chinese coal market, thus addressing a gap in the economic applications of this method. The model is used to analyze the sensitivity of factors influencing coal prices in China. The GPR–Sobol model effectively handles nonlinear relationships, enabling an in-depth exploration of key factors affecting price volatility and quantifying their impacts, thus overcoming the limitations of traditional econometric models in nonlinear data processing. The results indicate that economic growth, energy prices, interest rates, exchange rates, and uncertainty factors exhibit high sensitivity and significantly impact coal price fluctuations in China. Coal prices in northwest China are more sensitive to interest rates and geopolitical risks, while prices in east and south China are more responsive to exchange rates but less so to economic policy uncertainty. Additionally, coal prices in north, south, and east China are highly sensitive to international energy prices, indicating that coal prices are dominated by the global energy market, yet their weak response to macroeconomic indicators suggests these regions is still insufficiently mature.

1 January 2025 in International Journal of Energy Research
Verified
Journal Article
Open access
JLJingYe Lyu
JingYe Lyu
CLChong Li
Chong Li
The article proposes a feature selection framework that integrates principal component analysis (PCA) and random forest (RF) to identify the key factors influencing fluctuations in China’s new energy prices. Based on this, a parallel optimization comparison mechanism is constructed by integrating the enhanced whale optimization algorithm (EWOA), genetic algorithm (GA), and Newton–Raphson butterfly optimization (NRBO) algorithm for independent optimization of key multivariate variational mode decomposition (MVMD) parameters, effectively enhancing the performance of modal decomposition. Subsequently, the sample entropy (SE) method is introduced to perform modal reconstruction and frequency band division of the decomposition results. Based on the characteristics of each frequency band, autoregressive integrated moving average (ARIMA), bi‐directional gated recurrent unit (BiGRU), and bi‐directional long short‐term memory (BiLSTM) models are constructed, respectively, to fully exploit multiscale temporal information. Finally, a gradient boosting regression tree (GBRT) is used to integrate the outputs of multiple models for the final prediction. The experimental results show that the proposed method demonstrates significant superiority in the task of new energy price forecasting, achieving a mean absolute percentage error (MAPE) of 4.87%, the lowest among all benchmark models. It significantly outperforms existing comparison models, demonstrating high predictive accuracy and stability, and provides effective support for energy market modeling and decision‐making.

9 December 2025 in Sustainability
Verified
Journal Article
Open access
PDF file
SDShujuan Du
Shujuan Du
NLNa Li
Na Li
CLChong Li
Chong Li
JLJingye Lyu
Jingye Lyu
With the growing financialization of energy markets, financial and energy security have become critical global concerns. This study overcomes the limitations of traditional methods in analyzing extreme events by adopting a conditional quantile spillover index approach. Using China’s energy market prices and financial sub-market pressure indices, it constructs Quantile Vector Autoregressive (QVAR) models for both traditional and new energy-finance systems to examine their time-varying risk spillovers. Key findings are: (1) A significant risk spillover effect exists within China’s energy-finance system. The energy market acts as the primary risk transmitter, driven by both industrial policy and market demand, while capital and foreign exchange markets are the main risk absorbers. (2) The system exhibits significant tail spillover and asymmetry. The traditional energy market is more sensitive to upside extreme risks, whereas the new energy market is more sensitive to downside extremes. (3) Uncertainties like supply demand imbalances, policy shifts, and changing domestic/international conditions are major volatility drivers. Supply demand issues primarily affect the traditional energy market, while policy adjustments trigger chain reactions in the new energy sector. Based on these insights, the paper proposes recommendations to prevent systemic risks and potential energy crises.