期货市场量价分析法的短期预测力检验——基于久期理论的分析

久期作为重要的市场微观结构信号,对于改善资产价格波动预测准确性和流动性风险评价具有重要的作用。本文首先提出了反映价格、成交量和持仓量共同变化的共同久期概念,并分析了不同变量高频数据所具有的日内效应特征。然后,在久期理论框架下构建了扩展的二元选择模型,对上海燃料油期货市场量价分析法的短期预测力进行了实证检验。结果表明,期货市场量价分析法中一半以上的经验法则证明是有效的,说明量价分析法在短期价格预测中具有较高的预测力,是值得信赖的重要技术分析方法之一。     

Earnings dynamics in Canada: an econometric analysis

This paper reports the results of an empirical analysis of earnings dynamics in the Canadian labour market based on earnings data based drawn from tax returns between 1982 and 1994. Individuals' movements up and down quintiles of the earnings distribution are analysed using a hazard model approach. This represents one of the first studies for any country which models mobility across the entire distribution of earnings, including the middle and upper ranges as well as the lower ranges. The effects on transitions between quintiles are analyzed for the following variables: elapsed time spent in a given quintile (i.e., duration dependence effects), age, sex, geographical region, area size of residence, family status, language and overall macroeconomic conditions. One principal finding is that the conditional probability of transiting up or down the earnings distribution depends negatively on the elapsed time that an individual has spent in a given quintile. The earnings mobility patterns appear to be cyclical and exhibit some tendency of reversion to the mean, whereby the conditional probability of upward (downward) mobility is higher for those individuals presently situated in the lower (higher) quintiles.     

基于交易量持续期的流动性研究

本文充分利用我国限价指令驱动市场分笔数据所包含的信息。在交易量持续期的基础上，提出一个符合限价指令驱动市场特征的流动性指标，并从市场微观结构理论出发．选取了非对称信息的若干代理变量，分析非对称信息对市场流动性的影响程度。对万科A、B股的实证结果发现：（1）交易量持续期拥有信息含量，看涨行情会导致交易强度增大，看跌行情则导致较长的持续期。这反映了卖空限制的作用和投资者的追涨行为。（2）非对称信息是影响流动性水平的重要因素，投资者看法差异严重、价格剧烈波动以及长的交易量持续期导致市场流动性降低。     

东亚国家钉住汇率制度的可持续性研究

本文运用生存分析法，对东亚9国近60年钉住汇率制度的可持续性进行了实证研究。结果表明，东亚国家维持稳定的钉住汇率制度的总“生存时间”至多为51年，其影响因素主要是国内经济变量，故一国若要实现钉住汇率制度的维持与存续，需要保持相对较低的物价和利率水平，以及相对较高的货币供给增长率和国内信贷水平。     

Incidence and duration of unemployment spells: Implications for the male–female wage differentials

We demonstrate in this paper that the incidence of unemployment and duration of unemployment spells affect the male–female gender wage differential. The results indicate that after controlling for sample selectivity bias in the wage equation, the incidence of unemployment and duration of unemployment spells account for about 14.2% and 39.6%, respectively, of the wage differential that exists between males and females. We observed that while the duration of unemployment accounts for a substantial portion of the wage differential in both the service (48.8%) and the manufacturing (17.0%) sectors, the impacts of the incidence of unemployment and duration of unemployment spells variables on the wage differential are rather small in the service (0.07%) and the manufacturing (1.6%) sectors. Also, high incidence of unemployments is associated with low wages for males and females in all samples. In contrast, longer durations of unemployment spells tend to increase males’ wages and decrease females’ wages in the full and service sector samples. At the macro level, the results seem to suggest that promotion of gender wage equality should be associated with policies that will minimize the female's incidence of unemployment and duration of unemployment spells.     

New small firm survival in England

This paper investigates the determinants of the survival, between 2001 and 2004, of 622 small firms in England. Seventy one percent of these firms were less than 5 years old in 2001. Prior work by industrial economists has primarily focussed upon factors such as profitability and exit barriers. In contrast, this paper adopts a more managerial approach by examining whether the human capital of the business owner and organisational variables explain survival and non-survival. Our results suggest the founder’s education and bank finance promote firm survival. Firms which compete on price, or report being financially constrained at start-up, are much less likely to survive.

Large shareholders’ shareholding and earnings’ longitudinal persistence

This paper is the first to adopt longitudinal data analysis methods, such as the Kaplan-Meier survival analysis model, hierarchical linear model and Cox proportional hazards regression model, to investigate the relationship between large shareholders’ shareholding and earnings persistence. We find that large shareholders’ shareholding has a negative effect on corporate longitudinal persistence. The large shareholders’ shareholding will increase the risk of earnings decline and strengthen the declining trend of earnings. The effects of large shareholders’ shareholding on earnings’ longitudinal persistence for SOEs are different from those for non-SOEs, the effects in the pre-crisis period are also different from those in the post-crisis period. These findings contribute to the literature by adopting longitudinal data analysis methods and present new empirical evidence on the economic consequences of large shareholders’ shareholding.     

出口价格、出口品质与贸易联系持续期

贸易联系持续期对于一国出口贸易的平稳发展具有重要的意义。本文基于1996~2013年中国文化贸易HS6位码数据,通过构建出口品质、出口价格以及贸易联系持续期指标,采用多种回归模型实证分析发现以下结论。文化产品出口品质有助于文化贸易联系持续期的稳定,而出口价格则抑制了贸易联系持续期,尤其在核心文化产品种类以及较短的贸易联系持续期样本上。本文一个有趣的发现是,对于出口品质较高或贸易联系持续期较为稳定的文化产品样本,出口价格的负向抑制效应得到了极大的缓解,甚至产生正向促进效应。多种稳健性回归结果也均验证了上述结论。本文研究成果不仅为企业出口“价格-品质”争论提供了更为可靠的经验证据,同时对文化贸易联系持续期的稳定以及国际竞争力的提升具有一定的政策启示意义。     

Modeling trade duration in U.S. Treasury markets

This paper models the trading intensity of the US Treasury bond market, which has a unique expandable limit order book that distinguishes it from other asset markets. The results indicate that trade duration exhibits significant clustering and that the time taken to expand the tradable volume, known as ‘workup’, significantly decreases the time between the initiation of consecutive trades. Finally, we find that trade duration falls in the presence of scheduled news releases, but the size of the surprise in that news release is not found to be important.     

基于持续期数据的我国住房抵押贷款违约和提前还款风险分析

本文利用我国住房抵押贷款持续期数据,估计了住房抵押贷款在整个持续期内的提前还款风险和违约风险的概率,同时研究了各影响因素对住房抵押贷款这两类风险的影响方向。目前在国内文献中尚未见诸有关住房抵押持续期的实证研究,本文的研究能够反映抵押贷款风险的内在时间效应,这是其它统计分析方法无法做到的。本文进一步指出它们在银行的抵押贷款证券化和信贷风险管理中的应用前景。