全球股票市场系统性风险传递网络研究

本文采用系统性风险度量新指标LASSO-ΔCoVaR,构建全样本时期及各极端时期全球股票市场系统性风险传递网络,考察全球股票市场系统性风险传递水平及结构特征,并着重对极端状态下的风险传递进行分析。研究发现:第一,无论风险输入水平还是风险输出水平,不同股市的动态变化趋势大体一致,但波动幅度迥然不同,且单个股市风险输出水平的波动幅度远大于风险输入水平;第二,成熟经济体经济基本面恶化往往会增强其股市的系统性风险贡献,而新兴经济体则不同;第三,法国、荷兰、中国香港、德国和英国股市的风险溢出水平较高,同其他股市间的风险传递途径较多,是系统性风险传递网络中的核心节点;第四,我国股市与全球股市间的风险关联较弱,但我国股市潜在风险来源面广,同区域股市及金砖国家股市在我国股市与全球股市间的风险传递发挥重要作用。     

Does program trading contribute to excess comovement of stock returns?

Daily returns of stocks with high program trading comove more with each other but less with others. This significant comovement is disconnected with market movements and news of fundamentals and becomes stronger when market uncertainty is higher. It can be explained by neither the hypotheses of gradual information diffusion and liquidity provision nor the effects of quantitative trading signals, earnings announcements and index fund trading. Its non-fundamental nature is further demonstrated by the observation of program trading stimulating return reversals. Underlying this comovement is the high persistence of program trading. Our findings support the theory of habitat investing and demonstrate program trading creates a distinct source of excess return comovement.     

基于时间序列极值理论的跳跃风险研究

金融资产的跳跃行为作为对极端事件的刻画,为研究极端事件风险提供了良好工具。基于时间序列下的极值理论,在放松独立同分布假设下,构造了金融资产收益率序列尾部中跳跃动态特征的极值模型。通过对上证综指大跨度、高频度的实证研究,剖析了投资者结构、投资者行为与收益尾部分布之间的相互作用机制,进一步对金融资产收益尾部的跳跃风险进行了有效测度。结果表明,极端跳跃风险的分布特征在频率与尾部方向上呈现很强的不对称状态。     

基金投资风格漂移识别——基于收益和风险双维度

修正EGARCH-M模型构建基金投资风格漂移识别模型,考量收益和风险两个维度。实证发现,在较长时期内,中国基金投资风格不存在严重的漂移现象不明显,但在较短时期内,基金投资风格没有表现出较大的漂移度,相对于股市上涨阶段,股市下跌阶段基金投资风格发生漂移的概率更高。与现有的两种主要基金投资风格漂移识别方法相比较,模型具有四个方面的优越性,实证研究表明模型是可行的。     

Using possibility functions for long-term environmental planning

Current forecasting methodologies are deficient because they are only reliable for a few years into the future and focus on predicting the future. Serious environmental problems may require planning horizons in excess of 100 years. This article presents a planning method which uses possibility functions. Possibility functions that describe demographic variables can be combined with themselves and with probability functions that describe uncertainty about scientific knowledge. The article illustrates how possibility functions could be used to study the CO₂ problem.     

商业银行风险偏好设置与传导——基于巴塞尔协议视角的研究

风险偏好是银行在实现战略目标过程中愿意且能够承担的风险数量和种类,实质上是银行战略在风险管理上的具体体现。本轮金融危机表明,构建良好的风险偏好框架是建设稳健的全面风险管理体系的重要内容之一。风险偏好设置包括选取指标、量化指标值等;传导则包括自上而下的分解和自下而上的反馈过程。风险偏好应成为全程风险管理的主线。本文从风险偏好的定义、作用、设置和传导角度做一分析,以期对国内商业银行构建和实施科学的风险偏好框架有所启示。     

Public involvement in flood risk management in Hungary

The increasing frequency and intensity of recent floods and their economic, social, and political impacts created a situation in Hungary where flood management strategies need urgent reconsideration. Research suggests that in the case of natural disasters of uncertain and changing character, public education, information, participation, and cooperation are particularly important tools for coping. Drawing on the data collected by two recent empirical studies, this paper investigates state-of-the-art public involvement in flood control activities in Hungary. Results of a questionnaire survey and a series of semistructured interviews conducted in three flood basins of the Tisza river indicate that although the highly centralized system of flood control, and especially its strong financial background, a characteristic of state socialism, has significantly weakened since the political transition, forms of public participation, which could reduce the risk by building on a more conscious and responsible attitude of the citizens, have not developed yet. Paternalist and elitist attitudes prevail on the part of the authorities, contributing to the passivity of the public. There are signs, however, that in certain places, local government leaders take responsibility for building communication networks to raise public awareness and mobilize the public more effectively.     

Normality of stock returns with event time clocks

We introduce a different way to measure time using event clocks, with which we can observe a normal distribution of intraday stock returns. Most finance studies employ a ‘default’ time measurement that uses a calendar clock. Cumulative evidence from prior literature shows that returns with a calendar clock follow a distribution with an excess kurtosis and a heavier tail, relative to a normal distribution. We examine the distribution of intraday stock returns using different clocks. We find that returns do not follow a normal distribution with a traditional calendar clock, but do follow a normal distribution when event clocks are applied.     

典型事实、极值理论与金融市场动态风险测度研究

本文在金融市场典型事实约束下,运用ARFIMA模型对金融市场条件收益率建模,运用GARCH、GJR、FIGARCH、APARCH、FIAPARCH等5种模型对金融波动率进行建模,进而运用极值理论(EVT)对标准收益的极端尾部风险建模来测度各股市的动态风险,并用返回测试(Back-testing)方法检验模型的适应性。实证结果表明,总的来说,FIAPARCH-EVT模型对各个市场具有较强的适应性,风险测度能力较为优越。进一步,本文在ARFIMA-FIAPARCH模型下,假定标准收益分别服从正态分布(N)、学生t分布(st)、有偏学生t分布(skst)、广义误差分布(GED)共4种分布,对各股市的动态风险测度的准确性进行检验,并和EVT方法的测度结果进行对比分析。结果表明,EVT方法风险测度能力优于其他方法,有偏学生t分布假设下的风险测度模型虽然略逊于EVT方法,但也不失为一种较好的方法;ARFIMA-FI-APARCH-EVT不仅在中国大陆沪深股市表现最为可靠,而且在其他市场也表现出同样的可靠性。     

中国股票市场风险度量的模型选择与比较研究

分别采用等权移动平均方法、指教加权移动平均方法、GARCH(1,1)方法、GARCH(1,1)-t方法和Pareto型极值分布方法计算上海和深圳股票日收益率的VaR.向后检验表明,Pareto型极值分布方法比其他方法更能准确地反映我国股市的风险.     

政府在应对巨灾风险中的角色定位

当今世界,巨灾事故频发,各国政府为应对巨灾风险相继建立了巨灾风险保障制度,但是这些制度在运营过程中出现了一些问题,使得政府的投入与预想的效果相差甚远.本文重点分析了政府在参与应对巨灾风险组建巨灾保障制度的过程中,常用手段的利弊及遭遇的问题,以此来探讨政府在对应巨灾风险中应确立的合适角色定位.     

货币政策对不同类别股票影响的实证检验——基于中国台湾股市的数据

经实证检验发现,中国台湾实施的货币政策对所有类别股票的收益率及风险程度均有显著的影响,而且随着时间的流逝,这种影响有逐步加强的趋势;在货币紧缩时期,不同类别的股票受到的影响差异不大,在货币扩张时期,大盘股和成长股受到的影响大于其他类别的股票。     

Testing for differences in the tails of stock-market returns

In this paper, we use a database consisting of daily stock-market returns for 20 countries to test for similarities between the left and right tails of returns, as well as across countries. We estimate and test using the distribution of extreme returns over subsamples approach. Via Monte-Carlo simulations, we show that maximum-likelihood estimators are essentially unbiased, provided the size of subsamples is correctly chosen, and that the likelihood-ratio tests on parameters characterizing the behavior of extremes are correctly sized. For actual returns, we find that left and right tails behave very similarly. Across countries, we find that extremes are located at different levels and that their dispersion varies. The tail index, characterizing large extreme realizations, is found to be constant within each geographical group. We verify that the perception that left tails are heavier than right ones is not due to clustering of extremes. The failure to detect statistical significant differences is likely to be due to the relative infrequency of large extremes.