The Hubei carbon emissions trading market presents significant price volatility driven by energy price fluctuations, macroeconomic conditions and policy changes. Accurate price risk measurement is critically important for market participants. This study adopts Value at Risk (VaR) and Expected Shortfall (ES) to quantify market risk, and constructs a set of DCS-type models by combining the dynamic conditional score framework with the skewed Student-t distribution. Model evaluation covers unconditional coverage test, conditional coverage test, dynamic quantile test, the Actual-to-Expected ratio, the mean and the maximum absolute deviation, quantile loss and FZ loss. Empirical analysis based on daily HBEA spot prices from 3 April 2014 to 4 December 2024 shows that: (1) The DCS-ST model provides better data fitting performance and can effectively measure the market risk of China’s carbon trading market. (2) The parameter updating frequency has little impact on the prediction accuracy of the model. The results enriches the quantitative methodology for carbon market risk measurement and provide a reliable technical scheme for tail risk management in China’s carbon emissions trading market.

Effective risk management contributes to the stability of financial markets. This study investigates how decomposed investor sentiments affect financial volatility, emphasizing the role of sentiment polarity. Using BERT to extract positive and negative sentiment indices from stock forums, we find that negative sentiment has a stronger effect on realized volatility than positive sentiment, reflecting asymmetric investor reactions. Integrating these decomposed sentiment indices with technical indicators into a Transformer model significantly improves the forecasting of both realized volatility and value-at-risk (VaR). Our findings underscore the economic relevance of distinguishing sentiment polarity and offer novel insights into combining fine-grained sentiment features with deep learning for financial risk forecasting.

Influenced by energy prices, macroeconomic factors and policy factors, the price of the European carbon emissions trading market has changed significantly. Effective measurement of price risk of the European carbon emissions trading market is of practical importance for market participants. Two measures Value at Risk (VaR) and Expected shortfall (ES) are used to assess price risk. Based on the dynamic score (DySco) model and skewed Student-t (SKST) distribution, the DySco-SKST model is constructed and used to predict the price risk. The unconditional/conditional coverage tests, dynamic quantile test, Actual over Expected ratio, mean and maximum Absolute Deviation, quantile loss and FZ loss are used to evaluate the VaR and ES prediction performance. The daily spot closing prices of EUAs from January 3, 2013 to November 23, 2018 are used. The empirical results show that the VaR and ES prediction performance of the DySco-SKST model is better than that of the DySco-N and DySco-ST models. The VaR and ES prediction performance is affected by the parameter re-estimation scheme, but not by the parameter re-estimation frequency. The DySco-SKST model is better at predicting VaR and ES under the rolling window than under the expanding window.

对电力客户进行评价和细分,有助于电力企业提高差异化服务水平.首先从购电贡献价值、经营状况价值、信用价值和发展潜力四个方面构建电力客户价值评价指标体系,并提出利用理想物元可拓模型对电力客户价值进行评价;其次利用层次分析法和熵权法相结合的主客观集成赋权法来确定指标权重;接着利用正负理想物元和贴进度来替代经典域和最大隶属度准则,进而对待评估对象和正负理想物元的距离进行计算,得到了电力客户价值的综合贴近度及其完全排序;最后利用5家电力客户数据进行算例分析,同时针对研究结果给出了详细分析和客户维系策略.

In recent years, influenced by political and economic events, the price of the Shanghai crude oil futures market has changed significantly. It is therefore of great academic and practical importance to accurately measure the price risk of the Shanghai crude oil futures market. This paper uses a variety of GARCH models to predict price risk and uses the Model Confidence Set approach to evaluate forecasting performance. The daily closing prices of the Shanghai crude oil futures market from March 2018 to February 2021 are used. The empirical results show that futures price responds more strongly to negative news shocks than to positive news shocks, and the EGARCH model can effectively improve the accuracy of price risk measurement. An accurate assessment of the price risk can help investors to arrange funds in advance or to rebalance trading positions in order to meet the margin requirements.

In recent years, the frenetic advances of blockchain techniques have promoted the large-scale application of cryptocurrency and attracted significant attention in the mushrooming applications of decentralized finance (DeFi). To guarantee the health of a DeFi ecosystem, it is critical to reduce the transaction risks in a DeFi system. In particular, as a representative DeFi ecosystem platform, Ethereum's transaction process is mainly carried out with the help of smart contracts. Due to (pseudo)anonymity, the transaction process of Ethereum users is challenged by severe fraud threats. Ponzi scheme is the typical one. Previous studies have used machine learning methods to build Ponzi scheme detection models based on learning from the identified static smart contract samples feature data. However, in the early stage of smart contract deployment, the Ponzi scheme is difficult to detect. With the progress of transactions, Ponzi scheme will gradually show its characteristics. The existing methods are still falling short in capturing the temporal features of smart contracts for detecting Ponzi schemes in the big data environment. The recognition rate of the current approaches needs to be further improved. In this paper, we propose TTPS, a Long Short-Term Memory (LSTM) Ponzi scheme detection method considering time series transaction information of smart contracts. TTPS considers both temporal account features and code features of smart contracts. Adaptive synthetic sampling (ADASYN) is employed to effectively extend the feature data of minority class Ponzi scheme small samples. LSTM is utilized to learn from the temporal feature data of Ponzi scheme samples for TTPS model training. Experimental results verify and demonstrate the effectiveness and efficiency of TTPS.

Sensitivity analysis is at the core of risk management for financial engineering; to calculate the sensitivity with respect to parameters in models with probability expectation, the most traditional approach applies the finite difference method, whereafter integration by parts formula was developed based on the Brownian environment and applied in sensitivity analysis for better computational efficiency than that of finite difference. Establishing a similar version of integration by parts formula for the Markovian environment is the main focus and contribution of this paper. It is also shown by numerical simulation that our proposed methodology and approach outperform the traditional finite difference method for sensitivity computation. For empirical studies of sensitivity analysis on an NPV (net present value) model, we show the approaches of modeling, especially for parameter estimation of Markov chains given data of company loan states. Applying our newly established integration by parts formula, numerical simulation estimates the variations caused by the capital return rate and multiplier of overdue loan. Furthermore, managemental implications of these results are discussed for the effectiveness of modeling and the investment risk control.

With its transparent and fast claims payment, parametric insurance has been widely used to insure nature-related risks such as earthquakes, floods and hurricanes. In 2014, earthquake parametric insurance was introduced to provide coverage for earthquake losses occurred in Yunnan Province of China. However, as a main limitation of parametric insurance, basis risk is inevitable. In this paper, a Bayesian spatial quantile regression model is proposed to reduce the basis risk of earthquake parametric insurance. The effect of earthquake hazard, risk exposure, and vulnerability on economic loss are analyzed and considered in the quantile regression model. Since risk exposure and vulnerability at the epicenter cannot be observed, they will be treated as latent variables in the quantile regression model. Bayesian approaches are applied, and spatial correlation is considered to construct the prior distributions for the latent variables. Earthquake losses in Yunnan Province from 1992 to 2019 are collected and analyzed by the proposed model and methods. The payment mechanism and the corresponding premiums of 16 regions in Yunnan Province are then calculated. The results show that the loss ratio is more reasonable than the current earthquake insurance, and the basis risk is then reduced.

文章提出了一个处理含有零回报率的金融数据的框架,即将零值视为缺失的观测值,并用log-GARCH模型对回报率序列数据进行建模,同时结合QMLE方法和期望最大化(EM)算法对含缺失观测值的log-GARCH模型进行无偏估计,得到的条件标准偏差参数估计值能最大限度地接近真实值,提高了估计效率.

The application of blockchain technology is growing rapidly, which has aroused great attention in the academic and industrial fields. Based on blockchain 2.0, Ethereum is a mainstream smart contract development and operation platform. The trading process of Ethereum users is facing a serious threat of financial fraud. In particular, the Ponzi scheme is a classic form of fraud. Relevant works have investigated the issue of Ponzi schemes smart contract detection on Ethereum based on machine learning approaches. Nevertheless, the detection approaches still fall short in dealing with the big data-space Ponzi scheme smart contract detection application based on the class-imbalanced training data. We propose PSD-OL, a Ponzi schemes detection approach based on oversampling-based Long Short-Term Memory (LSTM) for smart contracts in this paper. PSD-OL takes the contract account features and the contract code features together into consideration. Oversampling technique is utilized to fill the class-imbalanced Ponzi scheme smart contracts’ sample feature data. An LSTM model is trained by learning from the feature data for future Ponzi scheme detection. Experimental results conducted on the well-known XBlock dataset demonstrate the effectiveness of the proposed method.

选取上海原油期货和中证新能指数作为研究对象,以VAR模型为基础,分析了原油期货价格对新能源行业股价产生的引导作用,研究了上海原油期货与我国新能源行业股价的相关关系.研究结果表明:①上海原油期货价格的下跌会引起我国新能源行业股价的上涨,两者表现出反向变动;②上海原油期货价格对新能源行业股价的贡献率在不断提高,相互影响程度逐渐增强.

现有GARCH模型依赖于参数条件分布形式假设,依然不能有效刻画金融资产收益偏态厚尾特性,分位数回归能给条件分布提供更加全面的描述.在分位数回归和GJR-GARCH模型基础上建立分位数GJR-GARCH模型,并在贝叶斯框架下对模型进行分析;同时利用中国金融市场数据检验分位数GJR-GARCH模型在风险价值预测方面的实际效果.

本文以2005年4月25日至2019年4月9日的欧盟碳配额(EUA)期货的日收益率作为研究对象,分阶段探讨资产价格时变跳跃的问题.本文利用常数跳跃强度模型和其他三种ARJI类模型来刻画跳跃强度和幅度具有时变动态性特征,研究结果显示:ARJI-GARCH、ARJI-R2t-1、ARJI-ht的拟合效果更佳;第一阶段价格变化较不确定,波动影响持续性弱,跳跃幅度易受市场波动影响;第二阶段波动较为持久,对历史事件敏感,且跳跃频率较高,但跳跃幅度方差与波动率之间联系较弱;第三阶段的跳跃幅度方差对于GARCH波动率有最强的敏感性.最后,本文从政策制定、市场监管、金融工具、市场参与者四个角度对发展我国碳市场提出建议.

This paper investigates the valuation of American option by developing a new technique for solving the optimal stopping problem modeled with Brownian motion. In particular, applying a martingale technique, the value function of option is approximated by specific basis functions. Compared with the traditional approaches, our method is more efficient without losing the accuracy. We show both the theoretical analysis and numerical implements.

“银村直联”是将银行业务系统嵌入村级财务管理平台,提供村级账户管理、信息查询、资金支付、非现金结算及增值服务等业务的新型金融产品,在实践中表现了实时、高效、便捷、精准等特点.通过“银村直联”,实现了银行农村金融服务信息化,助推农村集体资产财务管理的再次飞跃.以江苏紫金农村商业银行与南京市700个涉农村社区开展的“银村直联”合作为案例,针对该金融服务产品的流程再造和功能重塑展开分析,剖析该金融产品创新的现实意义和应用前景.

In this paper, we developed a sparse Bayesian variable selection in kernel probit model for high-dimensional data classification. Particularly we assigned a correlation prior distribution on the model size and a sparse prior distribution on the regression parameters. MCMC-based computation algorithms are outlined to generate samples from the posterior distributions. Simulation and real data studies show that in terms of the accuracy of variable selection and classification, our proposed method performs better than the other five Bayesian methods without the correlation term in the prior or those involving only one shrinkage parameter.

In portfolio risk minimization, the inverse covariance matrix of returns is often unknown and has to be estimated in practice. Yet the eigenvalues of the sample covariance matrix are often overdispersed, leading to severe estimation errors in the inverse covariance matrix. To deal with this problem, we propose a general framework by shrinking the sample eigenvalues based on the Schatten norm. The proposed framework has the advantage of being computationally efficient as well as structure-free. The comparative studies show that our approach behaves reasonably well in terms of reducing out-of-sample portfolio risk and turnover.

本文构建了基于偏斜广义t (SGT)分布的广义自回归得分(GAS)波动模型,并运用滚动预测方法计算多头头寸和空头头寸的VaR,同时为分析不同模型的样本外VaR预测效果,我们利用Kupiec和Christoffersen的方法来对VaR预测结果进行返回检验.以上证十大行业指数为例进行实证分析,研究结果表明:SGT分布族中能够灵活地刻画肥尾和偏斜特征的三种分布(ST、SGED、SGT)在VaR预测中具有优越性.综合考虑VaR预测效果与计算负担,基于ST分布的GAS波动模型是一个相对合理的选择.

A “buy low, sell high” trading practice is modeled as an optimal stopping problem in this paper. Because its award function lacks sufficient smoothness, traditional free-boundary approach with solution in form of integral equations is not available. Therefore, we design a backward recursive algorithm computing the value function to determine the stopping boundary. Besides, a new PDE technique is developed to conclude the special cases with positive drift. Finally, groups of comparison tests are designed to investigate the model parameters setting as well as the feasibility and profitability of the trading strategy.

选择合适波动模型和概率分布成为影响VaR预测可靠性的重要因素,本文首次结合HGARCH模型和AST分布来对金融资产收益率进行建模,并将所构建模型用于VaR预测研究中.我们重点比较研究不同头寸和不同风险水平下HGARCH模型及其子类共20种GARCH族模型的VaR预测效果,并系统性研究HGARCH模型中两个波动非对称参数在波动非对称性刻画和VaR预测中的作用.研究结果表明,在样本内,AST分布下的非对称GARCH族模型具有更好的波动拟合效果、分布拟合效果和VaR预测效果;HGARCH模型的两个波动非对称参数虽然在理论上是互补品的关系,但实际建模效果类似,相互之间更接近替代品的关系.在样本外,AST分布下的非对称GARCH族模型在波动拟合、分布拟合和VaR预测方面的优越性有所下降;两个波动非对称参数的实际效果也近似为替代品,其中放缩参数相比位移参数更具优势.