Jining Wang

The sustainable transformation of manufacturing industries requires a better understanding of risk linkages between high-carbon and low-carbon sectors, particularly under changing low-carbon policy conditions. This paper employs the TVP-VAR-BK model and the frequency-domain spillover method to systematically investigate the risk spillover characteristics and evolutionary patterns between high-carbon and low-carbon manufacturing industries from both dynamic and static dimensions. On this basis, a low-carbon policy uncertainty index (LCPU) is constructed, and the TVP-SV-VAR model is further applied to investigate the dynamic responses of risk spillovers between high-carbon and low-carbon manufacturing industries to LCPU shocks. The results show that: (1) Risk linkages exist between high-carbon and low-carbon manufacturing industries and exhibit a hierarchical pattern along the industrial chain. In terms of overall risk spillovers, the Chemical Industry (CH) and the four low-carbon manufacturing industries act as net risk transmitters within the system, while the other three high-carbon manufacturing industries mainly serve as net risk receivers. (2) Risk spillovers between high-carbon and low-carbon manufacturing industries are predominantly short-term. In the short term, the Chemical Industry (CH) exhibits a strong net risk transmission effect, whereas in the long term, all four low-carbon manufacturing industries act as net risk transmitters. (3) There is a pronounced dynamic relationship between changes in low-carbon policy uncertainty and risk linkages across manufacturing industries. The effects of LCPU shocks are concentrated primarily in the short term and decay rapidly as the forecast horizon increases. Moreover, high-carbon and low-carbon manufacturing industries differ in both the magnitude of their responses and their dynamic response paths, exhibiting clear time-varying characteristics and industry heterogeneity. These findings provide empirical evidence for improving cross-industry risk management and enhancing policy coordination, offering practical implications for maintaining industrial stability and supporting sustainable manufacturing transformation.

As agricultural commodities become increasingly financialised, reliable soybean futures forecasts are important for risk management and market monitoring. This study examines the main No. 1 soybean futures contract listed on the Dalian Commodity Exchange using a multi-source daily dataset spanning from February 2015 to December 2025. Ten predictors cover domestic price-volume information, cross-market transmission, the macro-financial environment, and supply–demand fundamentals. To address potential look-ahead bias, the forecasting protocol is strictly causal: monthly fundamentals are lagged by one month before daily alignment; Chicago Board of Trade (CBOT) settlement information is matched only from dates strictly preceding the corresponding Dalian Commodity Exchange (DCE) date; Min-Max scalers are fitted on the training sample only; and wavelet denoising, when used, is performed separately within each 60-trading-day historical input window. Models are trained on target dates through 2022, validated on 2023–2024, and evaluated on the fully held-out 2025 sample. Five fixed random seeds, a random-walk benchmark, directional accuracy, and Diebold–Mariano (DM) tests are used to assess robustness. The results show that thediscrete wavelet transform–bidirectional long short-term memory with attention (DWT-BiLSTM-Attention) specification is sensitive to random initialisation and has significantly higher squared forecast loss than the random walk for the five-seed mean forecast. Among the neural-network specifications considered, the DWT-LSTM model provides the lowest mean root mean square error (RMSE) and the smallest RMSE dispersion across seeds, although its predictive accuracy is not statistically different from the random walk. The ablation results further show that causal DWT and the ten-feature specification do not deliver robust incremental gains, while the incremental effects of attention and bidirectionality are not statistically significant. These findings highlight the importance of leakage-controlled preprocessing, strong naive benchmarks, and repeated-seed evaluation in financial time-series forecasting.

From the executive risk preference–policy uncertainty interaction perspective, this paper constructs a CA-SHIRD model by leveraging complex network theory and cellular automata, and further analyzes the liquidity risk contagion mechanism and dynamic evolutionary characteristics in the high-carbon manufacturing industry. In summary, this paper yields several key findings: (1) An increase in infection probability intensifies liquidity risk contagion, while a higher transition probability mainly accelerates short-term risk exposure. A higher recovery probability mitigates the contagion process and increases the proportion of recovered enterprises, whereas a higher default probability also weakens contagion but leads to a larger proportion of defaulted enterprises. (2) The liquidity risk contagion threshold in the high-carbon manufacturing industry is positively correlated with policy sensitivity, policy target ambiguity, and policy change frequency; negatively correlated with green preference, long-term orientation, and policy coordination; and exhibits a U-shaped relationship with risk perception and an inverted U-shaped relationship with policy instrument predictability. (3) When policy sensitivity interacts with policy change frequency, it exerts a risk resonance effect; when policy target ambiguity interacts with green preference and long-term orientation, a risk hedging effect is generated; when policy change frequency interacts with green preference and long-term orientation, the risk amplification effect of policy change frequency dominates the risk mitigation effect of green preference and long-term orientation.

This study adopts the Conditional Autoregressive Value at Risk (CAViaR) model to measure the downside tail risks of high-carbon-emission industries. Based on the Time-Varying Parameter Vector Autoregressive-BK (TVP-VAR-BK) model, it investigates the downside tail risk spillover effects across these industries. Meanwhile, this study constructs the Carbon Financial Regulatory Policy Uncertainty (CFRPU) index, and applies the Time-Varying Parameter Vector Autoregression with Stochastic Volatility (TVP-SV-VAR) model to examine the dynamic association between changes in CFRPU and changes in the total connectedness index. The results from the empirical study demonstrate that: (1) For the six sampled high-carbon-emission industries, downside tail risk spillover effects exhibit notable time-varying patterns alongside clear frequency heterogeneity. Specifically, the petrochemical sector serves as a net receiver of risk shocks over short-term horizons, but switches to a net transmitter of risk spillovers in the long term. (2) Bidirectional downside tail risk spillovers are observed across all six industries, with pronounced heterogeneity across investment horizons. Under the baseline 30-day frequency partition, the longer-horizon component is larger than the shorter-horizon component and broadly follows the temporal pattern of total connectedness. (3) The posterior mean response of ΔTCI to a positive ΔCFRPU innovation is largest at the two-month horizon and approaches zero by the six-month horizon. However, the 95% posterior credible intervals include zero, indicating that the short-horizon association is suggestive rather than statistically conclusive. These findings provide useful evidence for monitoring cross-industry financial risk and maintaining a predictable regulatory environment during the low-carbon transition.

This study examines risk spillovers among China’s green financial markets using the TVP-VAR-BK model. An Artificial Intelligence (AI) technology attention index is constructed and combined with an AI industry development index to capture AI shocks. The TVP-VAR-SV model is then employed to examine how these AI shocks interact with risk spillovers among China’s green financial markets. The results reveal that: (1) China’s green financial markets exhibit noticeable risk spillovers, mainly driven by short-term risk spillover, with considerable heterogeneity in markets’ roles as net risk transmitters and receivers across frequency horizons. (2) Risk spillovers are highly time-varying, increasing during periods of geopolitical tensions, public health crises, and industrial policy adjustments, but weakening as external conditions stabilize. Risk spillovers are predominantly short-term, except during the rapid development of generative AI, when noticeable long-term effects emerge. (3) The posterior mean responses of risk spillovers among China’s green financial markets to AI technology attention and AI industry development shocks exhibit time-varying characteristics and frequency-dependent heterogeneity, with predominantly positive and dynamic patterns. However, the 90% posterior credible intervals include zero during some periods, and these findings should therefore be interpreted as indicative dynamic patterns rather than conclusive evidence of effects credibly different from zero.

This study employs a TVP-VAR-BK-DY framework to examine volatility spillovers between China’s financial markets and strategic metal assets. To capture retail investor sentiment, we construct a sentiment index using an LLM knowledge distillation framework. Building on this index, the analysis further incorporates economic policy uncertainty to investigate the joint effects of retail investor sentiment and economic policy uncertainty on cross-market volatility spillovers. The results show that: (1) Price movements in certain assets exhibit leading effects, while metals with stronger financial characteristics generate more pronounced spillover effects. (2) The spillover structure between China’s financial markets and strategic metal assets displays substantial heterogeneity across time horizons and frequency bands. In the 1–5-day frequency band, the stock market serves as a net transmitter of volatility to the banking sector, gold, and copper. In the frequency band exceeding five days, these three assets exert reverse net spillover effects on the stock market. (3) The effects of retail investor sentiment and economic policy uncertainty on volatility spillovers differ significantly. The impact of retail investor sentiment is primarily concentrated in the 1–5-day frequency band, whereas economic policy uncertainty exhibits significant spillover effects in the frequency band exceeding six months.

The research employs principal component analysis to construct composite indices for China’s economic policy uncertainty (EPU), the green stock market, and the traditional stock market. Using nonlinear Granger causality tests and a Time-Varying Parameter Structural Vector Autoregression with Stochastic Volatility (TVP-SVAR-SV) model, it systematically examines the dynamic time-varying impact mechanism of China’s EPU on price volatility in the green securities and traditional financial markets. This research provides a crucial theoretical foundation and empirical reference, grounded in a nonlinear and time-varying perspective, for coordinating policy interventions to stabilize both traditional and emerging financial markets during the transition towards a green economy. The findings indicate: (1) An asymmetric risk transmission mechanism exists between the traditional financial and green securities markets in China, with the influence from the traditional financial market to the green securities market being stronger. (2) The influence of EPU in China on the green securities market exhibits time-varying characteristics that differ across periods. For instance, the Russia–Ukraine Conflict saw the green stock market’s most notable short-term negative response to EPU shocks. (3) The traditional financial market shows varied responses to China’s EPU shocks. China’s money market demonstrated a highly similar negative response pattern across three distinct periods.

The BP (Back-Propagation) neural network model (hereafter referred to as the BP model) often gets stuck in local optima when predicting China’s silicon wafer price, which hurts the accuracy of the forecasts. This study addresses the issue by enhancing the BP model. It integrates the principles of genetic algorithm (GA) with particle swarm optimization (PSO) to develop a new model called the GA-PSO-BP. This study also considers the material price from both the supply and demand sides of the photovoltaic industry. These prices are important factors in China’s silicon wafer price prediction. This research indicates that improving the BP model by integrating GA allows for a broader exploration of potential solution spaces. This approach helps to prevent local minima and identify the optimal solution. The BP model converges more quickly by using PSO for weight initialization. Additionally, the method by which particles share information decreases the probability of being confined to local optima. The upgraded GA-PSO-BP model demonstrates improved generalization capabilities and makes more accurate predictions. The MAE (Mean Absolute Error) value of the GA-PSO-BP model is 31.01% lower than those of the standalone BP model and also falls by 19.36% and 16.28% relative to the GA-BP and PSO-BP models, respectively. The smaller the value, the closer the prediction result of the model is to the actual value. This model has proven effective and superior in China’s silicon wafer price prediction. This capability makes it an essential resource for market analysis and decision-making within the silicon wafer industry.

This paper applies the Multivariate Stochastic Volatility (MSV) model, alongside its extended DGC-t-MSV model, and integrates Bayesian methods with MCMC techniques to develop the DGC-t-MSV-BN model. This model is specifically designed to analyze the volatility spillover effects between stock and futures markets. Key findings are as follows: (1) Significant volatility spillover effects exist from futures market to stock market. Notably, the spillover effects among the Chinese carbon futures market and both the Chinese and international stock markets are stronger than those within the Chinese carbon futures market, as well as the international gold and crude oil futures markets. (2) A notable negative volatility spillover effect is observed between Chinese carbon futures market and the international stock market. Conversely, a significant positive volatility spillover effect exists in the Chinese carbon futures market and the Chinese stock market. (3) The Chinese carbon futures market, as an emerging sector, displays high volatility and immaturity, yet it is developing at a rapid pace.

Based on the theoretical framework of Multivariate Stochastic Volatility (MSV), this paper combines the Dynamic Generalized Correlation (DGC) model with the t-distribution, establishes the DGC-t-MSV model, and employs the Markov Chain Monte Carlo (MCMC) algorithm based on the Bayesian principle for efficient estimation to investigate the price volatility spillover effects in China’s energy supply chains. The results of this study indicate the following: (1) The upstream crude oil spot price has a positive spillover effect on the midstream freight price. The downstream diesel market price, 92 gasoline market price, and 95 gasoline market price all exert positive volatility spillovers on the midstream crude oil freight price. (2) The volatility spillover effect between the upstream power coal price and the midstream coal freight price exhibits unidirectionality, and the volatility is transmitted from the power coal price to the coal freight price. (3) The upstream natural gas price and the midstream liquefied natural gas market price display asymmetric characteristics. Among them, the upstream natural gas price has a unidirectional and more pronounced positive volatility spillover effect on the midstream liquefied natural gas market price.