大数据背景下基于GRBF神经网络的P2P机构信用风险预警研究

针对P2P机构信用风险预警问题,提出了基于大数据思维的信用评估体系,采用基于动态特征的广义径向基神经网络对228家P2P机构12个月的高维数据指标进行信用风险评估.应用设计的广义径向基神经网络和BP神经网络进行对比,准确率分别为91.9%、85.2%,广义径向基神经网络在处理实时高维数据时表现出良好的性能,可以对我国P2P机构信用风险进行预警.同时深入对预警机构进行数据分析发现,如果企业资金流动性较差、净流入低也可能存在较高风险,企业应依据小额分散的借贷原则,降低借款集中度可以有效防范企业信用风险.     

海上通道关键节点风险预警研究

海上通道关键节点风险预警研究是保障海上通道安全的重要环节,为准确地预警关键节点的风险,通过对历史案例的统计分析,得出主要风险源因素,构建预警指标体系,以粗糙集理论和粒子群优化原理为基础,建立了基于最小网络误差的最优风险预警模型.通过实例分析证明了该模型可对海上通道关键节点的风险进行有效预警,为保障我国国际海运通道安全,降低海上运输风险和建设"21世纪海上丝绸之路"提供决策参考.     

基于内部时间的流动性风险度量模型研究

大额交易的流动性风险对投资者来说非常重要,已有模型更多从最优交易策略角度出发考虑这一问题,对市场特征的结合尚有不足.本文以内部时间概念为突破点,将其纳入已有最优交易策略模型,构建了新的流动性风险度量模型.结论显示,交易频率越高,市场流动性越好,流动性风险越小.     

流动性风险的度量

VaR模型忽略了流动性风险,到目前为止还没有统一的指标度量流动性风险.本文分析了最高成交价与最低成交价之差模型度量流动性风险存在偏差,同时给出了度量流动性风险一种新的修正模型.最后,结合实证对两种模型进行对比,修正模型从更加微观的层面上充分考虑到各价位的实际成交的价、量分布对总交易金额的作用,计算流动性风险值比较客观、精确.     

生猪市场价格风险预警的GRNN建模及其实证研究

为了对生猪市场价格风险进行预警,根据我国2009年1月-2011年8月14个指标的32组样本数据,建立了广义回归神经网络(GRNN)预警模型,其中训练样本29组,测试样本3组.训练样本和测试样本的均方根误差、平均绝对误差(AAE)和相关系数都非常接近,说明建立的模型具有较强的泛化能力和鲁棒性,测试样本的AAE为0.0062,平均相对误差为2.3%,说明建立的GRNN模型具有很高的预测精度,可用于我国生猪市场价格风险预警研究和实际预测,并为政府有关部门指导生猪生产和进行市场调控提供决策依据.     

GRNN和PNN模型在信用风险评估中的应用

广义回归神经网络GRNN和概率神经网络PNN,与传统的BP神经网络相比,收敛速度快,学习能力强.本文将其应用到信用风险评估,选取1057组公司财务数据作为训练数据,350组数据作为测试数据,分别建立基于不同属性的模型对样本公司财务状况评判其是守信公司还是违约公司,最终选取精度较高的作为最终模型对财务系统进行预测.结果表明,PNN对于信用风险评估泛化能力好,测试集正确率高,因此可以用作风险预警的模型,给决策者提供智力支持.     

基于W-SVM的民营环保企业信用风险预警模型

2019年中国绿色债券发行量依旧稳居世界前列,成为民营环保企业重要融资渠道,但是2018年至今,大量环保企业信用风险事件频发为我们敲响了警钟,构建合适的民营环保企业信用风险预警机制迫在眉睫.环保产业属于新兴产业,并以国有企业为主导,民营企业的样本数据具有样本量小,维度高等特征,这导致传统的信用风险模型适用性不强.因此选用加权支持向量机模型,对不同类别样本采取不同权值,选取大量财务特征,最终构建出风险预警模型．研究发现加权支持向量机模型具有十分优秀的预警性能.环保企业本身具有资金回收周期较长并且项目前期投入较高等特点,建议加强财务管理,保障资产流动性,建立完善产业链.     

基于流动性风险约束的我国商业银行资产负债随机规划模型

通过以资产负债管理合理匹配银行资产、负债,可以防范银行流动性风险.为此,建立了一个带有简单补偿的两阶段多期随机规划,在满足相关政策、法规约束和流动性风险V aR随机机会约束条件下,以银行的盈利最大化为目标,对银行主要资产、负债进行动态的优化匹配.     

流动性风险与股票收益率

文运用两阶段回归方法,利用上海证券交易所和深圳证券交易所的A股交易数据,针对流动性风险与股票收益率之间的关系进行了实证检验。在把流动性风险分解成系统流动性风险和个体流动性风险以后,作者发现无论是系统流动性风险还是个体流动性风险都对股票收益率有显著的影响;并且与成熟市场不同的是,在中国市场上个体流动性风险对收益率的影响比系统流动性风险的影响要更显著。因此在今后研究中国股市的流动性定价时,研究人员需要考虑个体流动性风险才能够对股票收益率进行更好的解释。     

上市公司财务预警的BP神经网络模型的建立及其应用

将BP神经网络方法应用于上市公司的财务预警上,构建了上市公司财务预警模型,不仅能发现企业是否存在风险和企业经营是否偏离轨道,向经营者提出警示,以便及时采取相应管理对策,而且还为广大的投资者和银行在内的债权人判定上市公司质量和经营业绩提供科学的手段和可靠的依据.实例分析表明该模型有效、可行,为上市公司财务预警提供了新的途径.     

Defensive financial decisions support for retailers in Greek pharmaceutical industry

In this paper, we present a decisions support solution designed for Greek pharmacies comprising a cash flow management system for early warning of financial distress and a financial advisor based on a neural network. The cash flow monitoring system integrates accounting elements with real time transactions and a predictive linear regression model while the decision support module is developed with the help of a neural network. For any given business unit the system associates accounting entries with information about credit times to reflect the precise instants of cash flows and using inflows/outflows equations monthly, eventually build its liquidity curve and cash flow balance over time. Alongside, a linear regression module is introduced to estimate future cash reserves based on past profitability ratios. Lastly, combining the power of artificial neural networks with expertise in this sector of pharmaceutical business, the financial decision support tool focuses on the retailers that face financial difficulties and suggests alternative solutions for escaping from distress and insolvency. The model has an ambitious and useful purpose, to inform and consult the owners of the business units and other members of the pharmaceutical chain, thus reduce financial risk for the chain.     

Filtering for risk assessment of interbank network

Our paper contributes to the recent macroprudential policy addressing the resilience of financial systems in terms of their interconnectedness. We argue that beneath an interbank market, there is a fundamental latent network that affects the liquidity distributions among banks. To investigate the interbank market, we propose a framework that identifies such latent network using a statistical learning procedure. The framework reverse engineers overnight signals observed as banks conduct their reserve management on a daily basis. Our simulation-based results show that possible disruptions in funds supply are highly affected by the interconnectedness of the latent network. Hence, the proposed framework serves as an early warning system for regulators to monitor the overnight market and to detect ex-ante possible disruptions based on the inherent network characteristics.     

Hedge fund systemic risk signals

In this paper, we realise an early warning system for hedge funds based on specific red flags that help detect the symptoms of impending extreme negative returns and the contagion effect. To do this we use regression tree analysis to identify a series of splitting rules that act as risk signals. The empirical findings presented herein prove that contagion, crowded trades, leverage commonality and liquidity concerns are the leading indicators for predicting worst returns. We not only provide a variable selection among potential predictors, but also assign specific risk thresholds for the selected key indicators at which the vulnerability of hedge funds becomes systemically relevant.     

ANN技术在企业经济预测与预警中的应用研究

提出了适合于上市公司而建立的基于ANN技术的企业经济综合指标短期预警系统的构建方案.该系统是一个人机相结合的反馈式预警系统,包括危机判定、财务指标预测、预警知识获取和报警四个子系统.该系统将定性分析与定量分析结合起来,既突出人的作用,又充分发挥了人工神经网络的在预测方面的技术,使得两者有机的结合在一起.其中,预警指标预测子系统体现了ANN技术在时间序列预测方面的应用,而知识获取子系统和报警子系统则体现了ANN技术在回归预测方面的应用,两者都有很好的理论基础.同时,该系统建立的程序比较规范,具有普适性,易于操作,比较容易实现.最后,还对该系统进行了实证模拟分析,并与专家意见结果进行了对比,验证了其有效性.     

A case study in applying neural networks to predicting insolvency for property and casualty insurers

This paper presents a neural network artificial intelligence model developed in cooperation with the Texas Department of Insurance as part of an early warning system for predicting insurer insolvency. A feed-forward back-propagation methodology is utilised to compute an estimate of insurer propensity towards insolvency. The results are then applied to a collection of all Texas domestic property and casualty insurance companies which became insolvent between 1987 and 1990 and the goal of predicting insolvency three years ahead of time. The results show high predictability and generalisability of results for the purpose of insolvency prediction, suggesting that neural networks may be a useful technique for this and other purposes.     

Indicator space configuration for early warning of violent political conflicts by genetic algorithms

Recognition of preconflict situations has a powerful potential for early warning of violent political conflicts. This paper focuses on the design and application of artificial neural networks as classifiers of preconflict situations. Achieving a desired level of performance of the neural network relies on the appropriate construction of recognition space (selection of indicators) and the choice of network architecture. A fast and effective method for the design of reliable neural recognition systems is described. It is based on genetic algorithm techniques and optimizes both the configuration of input space and the network parameters. The implementation of the methodology provides for increased performance of the classifier in terms of accuracy, generalization capacity, computational and data requirements. This revised version was published online in June 2006 with corrections to the Cover Date.     

Developing an early warning system to predict currency crises

The purpose of this paper is to develop an early warning system to predict currency crises. In this study, a data set covering the period of January 1992–December 2011 of Turkish economy is used, and an early warning system is developed with artificial neural networks (ANN), decision trees, and logistic regression models. Financial Pressure Index (FPI) is an aggregated value, composed of the percentage changes in dollar exchange rate, gross foreign exchange reserves of the Central Bank, and overnight interest rate. In this study, FPI is the dependent variable, and thirty-two macroeconomic indicators are the independent variables. Three models, which are tested in Turkish crisis cases, have given clear signals that predicted the 1994 and 2001 crises 12 months earlier. Considering all three prediction model results, Turkey’s economy is not expected to have a currency crisis (ceteris paribus) until the end of 2012. This study presents uniqueness in that decision support model developed in this study uses basic macroeconomic indicators to predict crises up to a year before they actually happened with an accuracy rate of approximately 95%. It also ranks the leading factors of currency crisis with regard to their importance in predicting the crisis.     

Risk application research on risk warning mechanism in organizational crisis management – taking Vanke Real Estate Co. Ltd., as an example

With the changes in domestic and international economic environment, the increasingly dynamic and complex environment has become the pressures and challenges that enterprises have to face. From the perspective of healthy development of companies in long-term running, it’s urgent to build an enterprise risk warning system.This paper takes daily operational risks and crises in Chinese enterprises as the research object, synthetically using the relevant knowledge of risk management theory, early warning management theory, the strategic management theory, the analytic hierarchy process, and fuzzy mathematics to build a daily management risk early warning system for Chinese enterprises. By constructing the system, the enterprises can make dynamic tracking for different stages in business management, so as to realize the risk before crises, take some actions during and after crises.     

基于BP神经网络的中国石油安全预警研究

为保障国家石油安全、保障国民经济安全运行,本文根据石油安全预警的特点,采用神经网络方法进行预警,设计了BP神经网络模型,应用BP神经网络建立了中国石油安全预警系统,并进行了预警研究和实证分析,得出我国石油安全属于重警区,需加强安全防范。