旅游安全的人工神经网络预警系统

针对于旅游安全预警问题,运用人工神经网络理论和方法,建立旅游安全预警的BP神经网络模型。通过实例进行分析,说明用BP神经网络方法是可行的。该模型具有很强的学习、联想和容错功能,其分析结果和过程都接近人脑的思维过程和分析方法,使得旅游安全预警结果的精度大大提高。     

基于混合粗糙集与ANN的上市公司财务危机预警研究

利用混合粗糙集ANN数据挖掘技术和相关敏感财务及非财务指标构建财务危机预警模型,对30家ST和90家非ST的A股制造业上市公司的财务数据进行实证检验。结果表明,将混合粗糙集的知识约简理论和神经网络的推理能力相结合构建的混合粗糙ANN预警模型具有较高的财务危机预测准确率和较好的实际应用价值。     

Incentivizing resilience in financial networks

When banks extend loans to each other, they generate a negative externality in the form of systemic risk. They create a network of interbank exposures by which they expose other banks to potential insolvency cascades. In this paper, we show how a regulator can use information about the financial network to devise a transaction-specific tax based on a network centrality measure that captures systemic importance. Since different transactions have different impact on creating systemic risk, they are taxed differently. We call this tax a systemic risk tax (SRT). We use an equilibrium concept inspired by the matching markets literature to show analytically that this SRT induces a unique equilibrium matching of lenders and borrowers that is systemic-risk efficient, i.e. it minimizes systemic risk given a certain transaction volume. On the other hand, we show that without this SRT multiple equilibrium matchings exist, which are generally inefficient. This allows the regulator to effectively stimulate a ‘rewiring’ of the equilibrium interbank network so as to make it more resilient to insolvency cascades, without sacrificing transaction volume. Moreover, we show that a standard financial transaction tax (e.g. a Tobin-like tax) has no impact on reshaping the equilibrium financial network because it taxes all transactions indiscriminately. A Tobin-like tax is indeed shown to have a limited effect on reducing systemic risk while it decreases transaction volume.     

Liquidity indicators,early warning signals in banks,and financial crises

This research compares the performance of three liquidity indicators, namely liquidity ratio (LiqR), liquidity creation (LiqC) and net stable funding difference (NSFD), for sending early warning signals for distressed banks. Recent evidence has shown that LiqR appears incapable of measuring the liquidity condition of banks. However, LiqC and NSFD have not yet been fully examined. Thus, which indicator is more useful in an early warning model becomes an interesting issue. We classify distressed banks as banks that have experienced a bank run, bailout, or failure. Sample data are collected from the United States and the European Union from before and after the financial crisis. We then estimate model predictive value using the sample before the crisis to predict liquidity shortages. Evidence shows that the academic (LiqC) and officially recommended indicators (NSFD) outperform LiqR as early warning signals. Furthermore, LiqC performs best when banks actively engage in income diversification but not fund diversification. Therefore, a well income-diversified bank with high LiqC tends to have high distress probability in the next period.     

基于BP神经网络的房地产预警研究——以沈阳为例

本文利用BP神经网络对房地产市场进行预警研究,选取了沈阳2005年²011年的预警指标数据,作为训练样本和检测样本进行仿真训练,其训练效果有着较好的可信度,该模型能够较好反应出警兆指标和房地产市场警情之间的关系,可以用于市场预警。     

高校网络舆情现状调研及对策研究——以南京航空航天大学为例

在高速发展的今天,高校和网络舆情已经有着密不可分的关系,尤其是校园网络有着自己独特的网络舆情。文章选取南京航空航天大学为研究个案,采用问卷调查的方法,分析网络舆论对大学生的影响状况和大学生对于网络舆论的基本态度,针对调查结果集中反映的问题提出相应的建议,给高校应对网络舆论带来的问题和危机提供参考。     

Real estate risk measurement and early warning based on PSO-SVM

The measurement and early warning of real estate risk are important to prevent and defuse major financial risks, and they form a basis for high-quality development. This paper assessed the internal and external environments of the real estate market; constructed a real estate risk indicator system from the aspects of market level, real estate enterprises, policy factors and financial institutions; and implemented a PSO-SVM model to measure and warn of real estate risk. Empirical studies were conducted. The results show the following: (1) the synthetic real estate risk index well depicts the cyclical fluctuation of real estate risk in Beijing; (2) the warning model based on the PSO-SVM method exhibits better performance and higher warning accuracy than other models do.     

Early warning indicators and macro-prudential policies: a credit network agent based model

Credit network configurations play a crucial role in determining the vulnerability of the economic system. Following the network-based financial accelerator approach, we constructed an agent based model reproducing an artificial credit network that evolves endogenously according to the leverage choices of heterogeneous firms and banks. Thus, our work aims at defining both early warning indicators for crises and policy precautionary measures based on the endogenous credit network dynamics. The model is calibrated on a sample of firms and banks quoted in the Japanese stock-exchange markets from 1980 to 2012. Both empirical and simulated data suggest that credit and connectivity variations could be used as early warning measures for crises. Moreover, targeting banks that are central in the credit network in terms of size and connectivity, the capital-related macro-prudential policies may reduce systemic vulnerability without affecting aggregate output.     

Analyzing the financial distress of Chinese public companies using probabilistic neural networks and multivariate discriminate analysis

Many studies have applied backpropagation feedforward neural networks (BPNNs) as an alternative to multivariate discriminant analysis (MDA) in attempts to predict business distress using relatively small data sets. Although these studies have generally reported the superiority of BPNNs vs. MDA, they seem to ignore the fact that the former suffers from overfitting if the data set is too small compared to the free parameters of the network. We thus suggest an alternative approach that involves use of a probabilistic neural network (PNN). From our study of financially distressed Chinese public companies, we found that both the PNN and MDA algorithms provide good classifications. Relative to MDA, however, the PNN method provides better prediction, and, at the same time, does not require multivariate normality of the data. Our results appear to offer an improvement from those of earlier efforts that employ MDA, BPNN, and other models. In particular, PNN was here able to predict company distress with greater than 87.5% short-term accuracy, and 81.3% medium-term accuracy.     

风险预警模型在房地产企业资金管理中的应用探讨

房地产企业资金管理普遍存在负债比例高、资金沉淀严重、投资行为盲目、融资形式单一、资金管理不善等问题。风险预警模型有预知风险、控制风险、防范风险等功能,运用风险预警系统对房地产行业进行监测、预测并判断其景气状态十分必要。选择了灰色风险预警模型和营运资金周转率作为房地产企业资金管理预警模型的核心指标,设计了房地产资金管理风险预警模型的具体算法和资金管理风险预警模型在房地产企业的具体应用。     

The study of contagious paces of financial crises

This study deals with the emergence of different regional crises and the comparison of early warning indicators to check for the accuracy of pace of exits. It was found that trade factors and monetary conditions clearly play a pivotal role in affecting the probability of existing time to currency crisis episodes and on the recurrence of crises. More specifically, using the index of market pressure methods, it is likely that the Asian Financial Crisis and the Mexico Tequila Crisis, when compared with the European Exchange Rate Mechanism (ERM) Crisis, were preceded by different spreads accelerating across those countries. The evidence suggests that efficient early warning indicators may exist and may be identified depending on the methods applied to the pace of exit involved.     

An agent-based early warning indicator for financial market instability

Inspired by the Bank of America Merrill Lynch global breath rule, we propose an investor sentiment index based on the collective movement of stock prices in a given market. We show that the time evolution of the sentiment index can be reasonably described by the herding model proposed by Kirman in his seminal paper “Ants, rationality and recruitment” (Kirman in Q J Econ 108:137–156, 1993). The correspondence between the index and the model allowed us to easily estimate its parameters. Based on the model and the empirical evolution of the sentiment index, we propose an early warning indicator able to identify optimistic and pessimistic phases of the market. As a result, investors and policy-makers can set different strategies anticipating financial market instability. Investors can reduce the risk of their portfolio while policy-makers can set more efficient policies to avoid the effects of financial instability on the real economy. The validity of our results is supported by means of a robustness analysis showing the application of the early warning indicator in eight different worldwide stock markets.

     

Systemic risk mitigation in financial networks

We propose a multi-period clearing framework, where the level of systemic risk is mitigated through the provision of liquidity assistance. The interbank liability network evolves stochastically over time, and assets of defaulted banks are sold to qualified banks within the network through a first-price sealed-bid auction. We find that policies targeting systemically important banks are more effective in core-periphery network structures, whereas those maximizing the total liquidity in the system are preferred in random network configurations. We assess sensitivity of systemic risk to variations in interbank liabilities as well as to their correlation structure.     

The default contagion of contingent convertible bonds in financial network

This paper investigates the impact of contingent convertible (CoCo) bonds on systemic risk using Eisenberg-Noe’s financial network method, in which the network is linked by debt relationships. As an efficient method for addressing the problem of “too big to fail,” CoCo bonds have received widespread attention, particularly because the trigger for CoCo bonds is a systemic risk event. Thus, the impact of CoCo bonds on systemic risk needs to be addressed. To solve this problem, we adopt default contagion and loss amplification due to network linkage to measure systemic risk, from which we can ascertain the potential impact on it of CoCo bonds. The results show that CoCo bonds enhance the spillover effect of the issuer’s default; meanwhile, sufficient CoCo bonds partly offset the impact of default contagion from other banks. Furthermore, CoCo bonds enhance the amplification effect of loss due to network linkage, but the amplification effect diminishes after the bankruptcy cost is considered. Finally, the numerical test provides some insight into how the issuance of writedown (WD) bonds influences commercial banks in China. Our study not only offers suggestions to the regulators of CoCo bonds but also contributes to related studies.     

Forecasting the realized volatility of Energy Stock Market: A multimodel comparison

The realized volatility forecasting of energy sector stocks facilitates the establishment of corresponding risk warning mechanisms and investor decisions. In this paper, we collected two different energy sector indices and used different methods, namely principal component analysis (PCA) and sparse principal component analysis (SPCA), to extract features, and combined LSTM and GRU to construct 12 different models. The results show that the SPCA-LSTM model we constructed has the best forecasting performance in the realized volatility forecasting of energy indices, and SPCA has better forecasting results than PCA in the feature extraction stage. The results of the robustness test indicate that our results are robust.     

浅析改善感知矿山物联网风险预警系统的研究

感知矿山物联网风险预警系统,主要是通过检测、分析矿山风险信息流,在事故萌芽阶段发出警报,以提示管理人员采取合理的消除隐患措施,来保障矿山生产安全。国内已有的矿山风险预警系统预警功能单一,准确率不高,仅仅起到报警的功能,无法体现预警的防控功能。为此,文章提出了一种改善矿山风险预警性能的方法,设计提高矿山风险的预知与预警,对提高矿山安全生产管理具有一定的现实意义。     

Systemic risk of Chinese financial institutions and asset price bubbles

Although there has not been a large-scale systemic crisis in China, high-risk financial events have occurred continuously in recent years. This research thus creatively analyzes the determinants of systemic risk for Chinese financial institutions from the view of asset price bubbles. First, we identify bubbles in the China stock and real estate markets on the basis of the generalized sup Augmented Dickey-Fuller (GSADF) model and explain the reasons for bubble formations according to the stage of China's economic development and policies implementation. At this stage, considering the differences in economic development levels of different cities, the real estate bubbles in the first, second and third tier cities and the whole country were innovatively identified. Second, on the basis of the DCG-GARCH-CoVaR model to measure the systemic risk of listed financial institutions in China and to classify institutions, the results show that the main source of such risk is the banking sector. Furthermore, by constructing regression models, stock market bubbles and real estate bubbles both positively correlate with systemic risk throughout the sample period. Meanwhile, the impact of bubbles on the systemic risk of different types of financial institutions was taken into account so that regulators prioritized different types of institutions with different characteristics when faced with decisions. Finally, we provide macro-prudential policy advice to regulators in order to weaken the impact of bubbles on financial stability to avoid systemic crises.     

工程项目投资风险评价模型研究

在项目决策前期对项目了解很少的情况下，一般对项目给出一个基本的风险综合评价数值，作为决策的依据。在建立工程项目投资风险评价指标体系的基础上，提出了将人工神经网络应用于工程项目投资风险评价中的综合评价方法，建立了人工神经网络模型，较好的解决了以往的评价方法中存在的人为确定权重的不足。最后进行了实证分析，在此基础上进行了敏感性分析，得到了比较好的评价结果。     

Systemic event prediction by an aggregate early warning system: An application to the Czech Republic

This work develops an early warning framework for assessing systemic risks and predicting systemic events over a short horizon of six quarters and a long horizon of 12 quarters on a panel of 14 countries, both advanced and developing. First, we build a financial stress index to identify the starting dates of systemic financial crises for each country in the panel. Second, early warning indicators for the assessment and prediction of systemic risk are selected in a two-step approach; we find relevant prediction horizons for each indicator by a univariate logit model followed by the application of Bayesian model averaging to identify the most useful indicators. Finally, we observe the performance of the constructed EWS over both horizons on the Czech data and find that the model over the long horizon outperforms the EWS over the short horizon. For both horizons, out-of-sample probability estimates do not deviate substantially from their in-sample estimates, indicating a good out-of-sample performance for the Czech Republic.     

自媒体环境下高校微博舆情干预机制的构建

随着自媒体时代的到来,传统的网络舆情也因为微博的广泛应用而表现出新的特征和规律。微博舆情具有主体去中心化、出现快速迅捷、形成依赖响应、扩散呈裂变式和发展可控性差的特点,其演变可简单分为萌芽、裂变、消解三个阶段。微博舆情的特点给高校传统的网络舆情处置工作带来了新的挑战,高校可通过完善信息公开长效机制、建立微博舆情监测机制、形成微博舆情应急处置机制和健全微博舆情反馈评测机制等方法构建适合自媒体环境的微博舆情干预处置机制。