Monetary policy and risk taking

We assess the effects of monetary policy on bank risk to verify the existence of a risk-taking channel – monetary expansions inducing banks to assume more risk. We first present VAR evidence confirming that this channel exists and is particularly significant on the bank funding side. Then, to rationalize this evidence we build a macroeconomic model where banks subject to runs endogenously choose their funding structure (deposits vs. capital) and risk level. A monetary expansion increases bank leverage and risk. In turn, higher bank risk in steady state increases asset price volatility and reduces equilibrium output.     

信用担保机构内部风险控制——基于COSO框架

我国中小企业信用担保机构是专门经营和管理信用风险的机构,其对资金的倍数放大作用同时也是对风险的放大。担保风险能否被有效地防范与控制关系到信用担保机构的可持续发展。信用担保机构的内部风险控制是防范担保风险的重要途径之一,文章结合COS0风险管理框架探讨了我国信用担保机构内部风险控制。     

国际资金循环测度的统计方法及应用

研究目标：本文研究如何测量国际资金循环（Global Flow of Funds,GFF）,试图整合国际金融相关数据源,创建GFF统计矩阵。研究方法：讨论了GFF的基本概念,确定了国际资金循环的统计范围。按照2008SNA所提倡的From Whom to Whom 的统计基准,建立了GFF的统计框架,据此框架整合了国际金融组织之间的数据系统关系和现有的国际金融统计数据源。研究发现：利用相关统计创建了GFF矩阵模型表,试编了一个包含中国在内的由11个国家所构成的“国家×国家”的GFF统计矩阵表。研究创新：创建了GFF矩阵表,展开了以中国为主要观测对象的国际跨境资本比较分析。研究价值：建立了观测GFF统计,开辟了新的分析视野,客观揭示了中国在GFF中的状况及现存的主要问题。     

信用担保机构内部风险控制——基于COSO框架

我国中小企业信用担保机构是专门经营和管理信用风险的机构,其对资金的倍数放大作用同时也是对风险的放大。担保风险能否被有效地防范与控制关系到信用担保机构的可持续发展。信用担保机构的内部风险控制是防范担保风险的重要途径之一,文章结合COSO风险管理框架探讨了我国信用担保机构内部风险控制。     

基于BP神经网络的拱坝风险评价研究

拱坝坝高库大,环境条件复杂,又多处于高烈度地震区;一旦失事,后果不堪设想。不但会严重影响工农业生产、重创国民经济,而且会使人民生命财产遭受极大损失,产生恶劣的政治影响。因此,拱坝的风险分析与风险评价越来越受到力学界、坝工界、政府的高度重视和社会各界的广泛关注。利用神经网络的方法,研究具有拱坝风险评价问题,建立了基于BP神经网络的评价模型。     

COMMUNICATION ABOUT FUTURE POLICY RATES IN THEORY AND PRACTICE: A SURVEY

We discuss the theoretical rationale for central bank communication about future policy rates, either as part of inflation targeting or in the form of forward guidance. We also summarize both actual central bank communication about future policy rates and empirical evidence on the effectiveness of these types of communication. We argue that there is a disconnect between the theory and practice of forward guidance, as theory assumes commitment on the part of the central bank, while in practice central banks generally do not commit. Future theoretical research on forward guidance should therefore take the absence of commitment by central banks into account.     

The composite iteration algorithm for finding efficient and financially fair risk-sharing rules

We consider the problem of finding an efficient and fair ex-ante rule for division of an uncertain monetary outcome among a finite number of von Neumann–Morgenstern agents. Efficiency is understood here, as usual, in the sense of Pareto efficiency subject to the feasibility constraint. Fairness is defined as financial fairness with respect to a predetermined pricing functional. We show that efficient and financially fair allocation rules are in one-to-one correspondence with positive eigenvectors of a nonlinear homogeneous and monotone mapping associated to the risk sharing problem. We establish relevant properties of this mapping. On the basis of this, we obtain a proof of existence and uniqueness of solutions via nonlinear Perron–Frobenius theory, as well as a proof of global convergence of the natural iterative algorithm. We argue that this algorithm is computationally attractive, and discuss its rate of convergence.     

Deep learning for modeling the collection rate for third-party buyers

This study evaluates a wide range of machine learning techniques such as deep learning, boosting, and support vector regression to predict the collection rate of more than 65,000 defaulted consumer credits from the telecommunications sector that were bought by a German third-party company. Weighted performance measures were defined based on the value of exposure at default for comparing collection rate models. The approach proposed in this paper is useful for a third-party company in managing the risk of a portfolio of defaulted credit that it purchases. The main finding is that one of the machine learning models we investigate, the deep learning model, performs significantly better out-of-sample than all other methods that can be used by an acquirer of defaulted credits based on weighted-performance measures. By using unweighted performance measures, deep learning and boosting perform similarly. Moreover, we find that using a training set with a larger proportion of the dataset does not improve prediction accuracy significantly when deep learning is used. The general conclusion is that deep learning is a potentially performance-enhancing tool for credit risk management.     

Combining forecasts for universally optimal performance

There are two potential directions of forecast combination: combining for adaptation and combining for improvement. The former direction targets the performance of the best forecaster, while the latter attempts to combine forecasts to improve on the best forecaster. It is often useful to infer which goal is more appropriate so that a suitable combination method may be used. This paper proposes an AI-AFTER approach that can not only determine the appropriate goal of forecast combination but also intelligently combine the forecasts to automatically achieve the proper goal. As a result of this approach, the combined forecasts from AI-AFTER perform well universally in both adaptation and improvement scenarios. The proposed forecasting approach is implemented in our R package AIafter, which is available at https://github.com/weiqian1/AIafter.     

Evaluating quantile forecasts in the M5 uncertainty competition

Probabilistic forecasts are necessary for robust decisions in the face of uncertainty. The M5 Uncertainty competition required participating teams to forecast nine quantiles for unit sales of various products at various aggregation levels and for different time horizons. This paper evaluates the forecasting performance of the quantile forecasts at different aggregation levels and at different quantile levels. We contrast this with some theoretical predictions, and discuss potential implications and promising future research directions for the practice of probabilistic forecasting.