摘要
本文旨在研究如何在金融风险测度中对关键性隐患——不确定性进行适度的量化分析以有效增强风险管理的审慎性。首先直观呈现“不确定性”的概率统计表现,分析其引致风险或危机事件的必然性与严重性。进而,以广泛使用的风险管理方法VaR与ES为例全面回顾与分析相应领域的技术发展历程,揭示解决不确定性问题的重要性。由此,基于概率统计领域的国际前沿突破与相应参数估计方法的创新发展,系统性提出兼容无穷可能不确定性分布的风险审慎管理模型GE-VaR与GE-ES,进一步地,通过与公认最有效的风险测度方法相比较,证实纳入概率分布不确定性的风险管理模型的敏锐性与审慎性,以及对中国现阶段高波动率、相对高风险、高不确定性市场特征的适用性。
According to Tversky & Fox (1995), decision theory distinguishes between risky prospects, where the probabilities associated with the possible outcomes are assumed to be known, and uncertain prospects, where these probabilities are not assumed to be known. However, most decisions are under both risk and uncertainty, which lie somewhere between the two extremes of decisions under risk and decisions under ignorance. Typically, decision makers do not know the exact probabilities associated with outcomes, but they have some vague notion about their likelihood. Especially in business situations, as Knight (1921) suggested, it is so rarely possible to quantitatively determine the true probability. The global financial turmoil in recent years again highlights the necessity and importance of managing uncertainty in business decision and economic forecasting. The purpose of this paper is to conduct a quantitative analysis of uncertainty and thus to improve risk modelling with uncertain probability distribution. The paper first conducts an empirical analysis to explore the uncertain characteristics of financial data. The paper then analyses how inevitably uncertainty plays an important role in the increase of financial vulnerability. Specifically, we show that it is difficult to predict which exact segment of financial time series a future log-return belongs to while different segments have quite different statistical attributes. It is also difficult to predict the statistical characteristics, including the shape, of the probability distribution of the possible segment of data. The uncertainty brings difficulty in risk management, especially when the financial market undergoes frequent and big changes. Motivated by the above observation and based on a comprehensive review of the development of Value at Risk (VaR) and Expected Shortfall (ES), the paper further demonstrates the importance of incorporating uncertainty in tail risk modelling. The review reveals that emphasis on the consistent improvement of accuracy of paramet
作者
宫晓琳
彭实戈
杨淑振
孙怡青
杭晓渝
GONG Xiaolin;PENG Shige;YANG Shuzhen;SUN Yiqing;HANG Xiaoyu(School of Economics, Shandong University;School of Financial Studies,Shandong University;School of Mathematics, Shandong University)
出处
《经济研究》
CSSCI
北大核心
2019年第7期64-77,共14页
Economic Research Journal
基金
泰山学者工程专项经费
山东大学杰出青年、青年学者未来计划资助
