证券投资基金收益概率密度预测——基于神经网络分位数回归模型

证券投资基金收益往往具有更高的峰度与更大的偏度,建立在古典假定基础上的均值回归分析难以给出准确预测结果。考虑到证券投资基金收益中的高峰、非对称等典型特征与各因素对收益序列的非线性影响模式,建立神经网络分位数回归模型,一方面,可以通过分位数回归功能,揭示各因素对证券投资收益整个条件分布的影响规律;另一方面,可以通过神经网络结构,模拟金融系统中的非线性关系。在神经网络分位数回归模型基础上,对证券投资基金收益整个条件密度函数进行预测,提供比点预测更多的有用信息,便于进行科学决策。     

Asymmetric quantile analysis of the Swedish mortgage price discovery process

Based on Swedish banking data we discover robust and significantly positive Asymmetric Price Transmission (APT) effects over all analysed regression quantiles of our mortgage interest rates, with even larger positive APT for the higher percentiles. The analysis was enabled through unique access to a Swedish bank's (SEB) own records of their true borrowing costs. Our central contribution is that there is a higher propensity for the bank to rapidly increase its mortgage interest rates for customers following an increase in its borrowing costs, compared with the propensity for the bank to decrease its customers’ mortgage rates subsequent to a corresponding borrowing cost decrease.     

The Exporter Wage Premium Hypothesis: An Unconditional Quantile Regression and Decomposition Approach

ABSTRACT

Relying on the World Bank Enterprise Survey dataset in 2012/13, this paper applies the unconditional quantile regression and decomposition estimation techniques to examine the hypothesis that workers in exporting firms receive higher wages than those in non-exporting firms. The results show that the relationship between export and firm’s wage bill is indirect and is transmitted through technology and firm size. Remarkably, these indirect relationships are much more pronounced at the more upper quantiles of the wage bill distribution. However, the net relationships of the interaction between export and technology are relatively larger and positive as compared to that of the interaction between export and firm size which are marginal and mixed. The decomposition analysis indicates that much of the present exporter wage premiums are largely due to the differences in the returns to the characteristics between exporting and non-exporting firms. The findings from this paper suggest directions for future work that can be directly useful for policy.     

VaR技术的统计学分析

目前，VaR技术已成为国际上度量市场风险的主流方法。本文从VaR技术的概率内涵出发，对不同概率分布下的VaR计算进行了统计分析，并在正态分布下推导出了求解VaR的统计学意义下的计算公式。最后。对VaR技术的要素进行了分析。     

Revisiting growth empirics based on IV panel quantile regression

We analyse the well-known issue of economic growth convergence using quantile regression. Most previous studies have used a least squares (LS) method or variation, which focuses on the issue only at the mean of the growth rate. Therefore, such results cannot provide a satisfactory answer to what can happen if the growth rate is far from the conditional mean level. For example, we consider the following question: do we still have economic growth convergence or is the convergence speed changed in a low growth period such as the ‘Great Recession,’ that started in 2008? We propose using instrumental variable panel quantile regression to answer this question. Our empirical findings demonstrate that economic growth convergence occurs at all quantiles over the entire conditional distribution, but that the convergence speed does depend on quantiles; the convergence speed is much higher when the GDP growth rate is at either high or low quantiles.     

CEO Pay-For-Performance Heterogeneity Using Quantile Regression

We provide some examples of how quantile regression can be used to investigate heterogeneity in pay‐firm size and pay‐performance relationships for U.S. CEOs. For example, do conditionally (predicted) high‐wage managers have a stronger relationship between pay and performance than conditionally low‐wage managers? Our results using data over a decade show, for some standard specifications, there is considerable heterogeneity in the returns‐to‐firm performance across the conditional distribution of wages. Quantile regression adds substantially to our understanding of the pay‐performance relationship. This heterogeneity is masked when using more standard empirical techniques.     

借款描述对互联网金融信用风险的影响研究

基于互联网金融提供的客户借款描述,通过潜在语义分析挖掘借款描述文本内容的主题,运用二元分位数回归分析借款描述对互联网金融信用风险的影响。实证结果表明,借款描述中有关情感表达、个人信用和借款目的的描述与违约情况呈负相关,有关财务情况的描述与违约情况呈正相关。     

Hedging performance analysis of energy markets: Evidence from copula quantile regression

This study investigates hedging performance with respect to different market structures for energy-related commodities, including West Texas Intermediate crude oil, Brent crude oil, Chinese crude oil, and Heating oil. Copula quantile regression functions and the generalized autoregressive conditionally heteroscedasticity model are combined to analyze the nonlinear impact of dependence and the heterogeneous impact of market structure changes on hedging performance. Results show that hedging performance presents nonlinearity and market structure changes have surprisingly strong heterogeneous effects on the quantile hedge ratio, where bearish and bullish have lower hedge ratios than normal markets, which is captured better by Clayton copula quantile regression. Additionally, the trend of hedging effectiveness over different market structures also shows an inverted U shape. After changing data frequency or the types of futures contracts, the conclusions remain the same. Our empirical findings imply that hedgers are supposed to adjust the hedging number of futures according to market structure changes to hedge price risk effectively.     

财政分权与市场整合对 工业用地利用效率影响的异质性研究

本文利用2005—2017年省际面板数据,运用SBM-DEA模型测算中国30省份（西藏地区和港澳台地区因数据缺失而剔除）工业用地利用效率。随后,构建面板分位数回归模型,研究财政分权与市场整合在不同分位点处对工业用地利用效率的异质性影响。研究发现：第一,当工业用地利用效率较低时（0.1~0.3分位点处）,财政分权对工业用地利用效率具有显著的负向影响（影响系数为－0.6402~－0.1998）;而随着分位点提升（0.4~0.9分位点处）,财政分权对工业用地利用效率将产生正向促进作用（影响系数为0.2301~0.7150）;财政分权对工业用地利用效率增长率的影响随着分位点提高大致呈倒“U”型变化趋势。第二,随着分位点的提高,市场整合对工业用地利用效率增长的贡献趋于平稳,对工业用地利用效率增长率的正向促进则逐渐减弱。第三,财政分权与市场整合的共同作用在低分位点处对工业用地利用效率及其增长率具有负向影响（对工业用地利用效率的影响系数为－0.0294~－0.0088;对工业用地利用效率增长率的影响系数为－0.2635~－0.2838）,在高分位点处则会产生正向促进作用（对工业用地利用效率的影响系数为0.0042~0.0540;对工业用地利用效率增长率的影响系数为0.2004~1.1496）。     

Nonlinear causality relationship between stock and real-estate returns in PIGS countries: wealth effect or credit-price effect

It is the first research to investigate for nonlinear interdependence of these two markets in the PIGS (Portugal, Italy, Greece and Spain) countries based on the quantile causality test. The results reveal the existence of the nonlinear causality relationship between the stock returns and real-estate returns in the PIGS countries.

The empirical results of the quantile causality test suggest a significant causal relationship between these two markets in the PIGS countries, especially in the tail quantile. The existence of a significant tail interdependence implies that investors are unable to hedge the risk across the real-estate and stock markets when they are extremely volatile. Therefore, when there exist extreme returns between the two markets in the PIGS countries, both continuous negative impacts imply that instability in the real-estate market drives instability in the stock market and vice versa. It could be one of the major reasons why it deepened the systemic risk of the European sovereign debt crisis.