基于Hawkes过程中美股市大幅波动互激效应的研究

近年来，由于中美经济联系日趋紧密，中美股票市场大幅波动的互激效应明显增强。本文考虑中美股市时差和法定节假日差异等因素，运用标值Hawkes过程对2006-2017年CSI300和S&P500大幅波动收益率数据进行建模，结果表明：（1）中美股市大幅波动互激效应存在不对称性，美股市场大幅波动对中国股市的互激效应更强；（2）中美股市大幅波动的幅度对互激效应不存在显著影响；（3）中美股票市场对于大幅波动互激效应的消化速度存在差异，中国股票市场消化美股大幅波动互激效应的速度较快。本研究对金融市场监管者和投资者均有一定意义。     

微信公众号影响力指数建构与量化评估

作为被广泛使用的社交媒体，微信已经成为热点事件和网络舆情的重要策源地，科学准确评估微信公众号的影响力对于明晰微信信息传播特点，深入挖掘微信传播潜力，进一步规范和指导微信平台的传播行为具有较大的理论和实践意义。在对既有关于微信公众号影响力研究进行梳理和评价的基础上，基于传播效果测量的设计，坚持以受众为中心、由浅入深多层次考察准则，运用德尔菲法和层次分析法建构一套微信公众号影响力指标体系。该指标体系由4个一级指标和9个二级指标构成，并运用线性合成法合成微信公众号影响力指数。     

公共卫生支出、健康结果与卫生投入政策——基于189个国家和地区的面板门槛分析（1995—2011年）

利用189个国家和地区1995—2011年的数据实证检验公共卫生支出规模对一个国家的健康结果（婴儿死亡率和出生时的平均期望寿命）的影响，结果表明，公共卫生支出占GDP的比重对婴儿死亡率和出生时的平均期望寿命的影响具有门槛效应，门槛值分别位于1.9%和6.62%。分段回归结果显示，公共卫生支出占GDP的比重低于1.9%时，公共卫生投入对出生时的平均期望寿命没有显著影响，对婴儿死亡率的影响较弱；超过1.9%后，公共卫生投入的规模效应开始体现，对健康结果指标的影响均显著增强；超过6.62%后，公共卫生投入对婴儿死亡率的影响不再显著，对出生时平均期望寿命的弹性系数不再变化，单位边际贡献不再增加。中国当前政府卫生投入规模仍然较低，需要继续增大公共资金投入、降低个人卫生支出比重。     

中国城镇居民家庭储蓄水平队列效应分析

利用中国家庭收入调查（ CHIP）数据对城镇居民家庭的储蓄率进行了队列效应的实证研究。分析结果表明，属于不同队列的城镇家庭在储蓄水平上都呈现出相似的“U”型年龄分布，即中年家庭（40－50岁）储蓄率低，年轻家庭（20－40岁）和退休家庭储蓄率较高，这一现象与其他国家的实证研究结果有显著不同。本文认为，中国城镇居民储蓄率的“U”型年龄分布主要是由人口与经济的高速增长带来的：人口高速增长带来的队列规模效应使得面对更大竞争压力的年轻人口储蓄率较高，而经济水平持续增长带来的消费水平增长预期使得接近退休家庭储蓄水平重新上升。     

技术创新与企业就业效应--基于微观企业调查数据的实证分析

利用2012年世界银行关于中国工业企业的调查数据，从流程创新、产品创新和企业研发三个维度分析了短期内技术创新对就业增长的影响。结果显示，流程创新与就业增长显著负相关；产品创新与就业增长显著负相关；企业研发与就业增长负相关，但较前两者显著性较弱。进一步研究发现，与东部沿海地区相比，中西部地区技术创新的就业替代效应更为明显。鉴于上述研究结果，对现有技术创新方式和政策提出相应的对策建议。     

小学校长服务型领导对教师知识分享行为的影响研究

以山东省248名小学教师为有效样本，探讨了服务型领导、教师同事信任、情感承诺和教师知识分享行为之间的关系。研究结果显示：服务型领导与教师知识分享行为显著相关；教师同事信任部分居间联系了服务型领导与知识分享行为之间的关系；教师同事信任部分居间联系了服务型领导与情感承诺之间的关系；情感承诺对服务型领导与知识分享行为关系的中介作用不显著。     

环境管制、知识产权保护与区域创新能力——基于我国省际面板数据的实证研究

文章基于政府主导视角,利用2006—2015年我国30个省级面板数据,构建多元回归模型,分析变量之间的关系.其结果显示,环境管制、知识产权保护与区域创新能力之间呈非线性关系,环境管制与知识产权保护同时实施对区域创新能力有显著的促进作用,且具有区域异质性.从企业生存角度出发,提出门槛效应假说,采用Hansen面板门槛回归模型进行检验.结果表明,在环境管制与知识产权保护共同发挥作用时,环境管制必须控制在合理的范围内.最后采用门槛效应回归方法找到了环境管制与知识产权保护对区域创新能力影响的门槛值,并提出相应的政策建议.     

基于超总体模型的设计效应分解

本文基于超总体模型研究抽样调查中设计效应的计算问题。首先以随机效应模型为基础，明确了简单随机、二阶段、不等概率和分层抽样对应的超总体模型，进而通过所给模型推导出分层、类集、加权单因素设计效应的计算公式和多因素组合的设计效应计算公式并给出了对应估计量，公式表明：多因素同时存在的组合设计效应等于对应单因素设计效应的乘积。最后，对设计效应的理论值、估计值和真实值之间的关系进行了蒙特卡洛仿真，并利用相对偏倚、相对均方误进行了评价。本文的研究，对复杂抽样设计中正确计算、使用设计效应具有指导意义。     

新农合对四川藏区的减贫效应分析

The New Rural Cooperative Medi-cal System ( hereafter NRCMS) in Tibetan areas of Sichuan was started in Wenchuan in 2005 , and by 2008 covered all of the province’s Tibetan areas. This paper studies the effects of the NRCMS on im-proving the health of and alleviating poverty for farmers and herdsmen in Tibetan area of Sichuan. Most parts of the Tibetan areas of Sichuan are located in high altitude districts. Thirty two coun-ties of these areas are classified as “National Pov-erty Counties”. Poverty and disease go hand in hand in these regions. Kashin-Beck disease and hydatid disease are the major endemics in the pas-toral and agro-pastoral areas of Sichuan. Endemic, infectious and chronic diseases are widespread in Sichuan’s Tibetan areas. More than 70% of pa-tients are workers from 20 to 60 years old. Disea-ses are more prevalent in women than in men. Kashin-Beck disease and hydatid disease are cur-rently incurable. Patients suffer from health prob-lems, which leads to a decrease in their income and the heavy burden of medical expenses. The new rural cooperative medical system alleviates the negative effects of farmers’ falling into, or back in-to poverty due to disease. However, the existing medical compensation mechanism is not sufficient to solve the problem. The greatest impact of NRCMS on the farmers and herdsmen in Sichuan’s Tibetan areas is that the system has gradually changed local people’s medi-cal behavior, as well as their underlying ideas a-bout medicine: they begin to believe in hospitals. In particular, more pregnant women are choosing to give birth in hospitals, which reduces the rate of infant mortality and postpartum diseases, and im-proves the health of women. Since the full coverage of the NRCMS in 2008 , the number of people participating in the system has reached the overall average level of Si-chuan province. By analyzing the data before and after the implementation of this system, and meas-uring the impact of the system on people’s health, it can be found that the NRCMS’s role in serving the vulnerable population, such as the elderly and infant children, is more marked. Since the implementation of the NRCMS, all administrative villages in Sichuan Tibetan areas have established village clinics, which solved the problem of a shortage of medicines and doctors in those areas. Farmers and herdsmen have conven-ient access to medical treatment, enhancing the ac-cessibility of medical service. After the implemen-tation of the NRCMS, the health of the elderly population in rural areas has improved. Infant mor-tality rates have dramatically fallen. The implementation of the NRCMS improved the medical service capacity of township hospitals and village clinics. And the NRCMS has brought the township hospitals and village clinics into its scope of compensation, which greatly promotes the utilization of primary medical services in Tibetan areas. The poverty reduction effect of the NRCMS can be analyzed from two aspects:Firstly, the im-
provement in health leads to increased income, be-cause good health can promote labor productivity. Meanwhile, the increase in income will in turn im-prove the overall level of health. Secondly, the in-patient and outpatient compensation rate is raised year by year, which reduces the medical fees of farmers, and prevents them from falling back into poverty.     

Conditional Akaike Information Criteria for a Class of Poisson Mixture Models with Random Effects

Focusing on the model selection problems in the family of Poisson mixture models (including the Poisson mixture regression model with random effects and zero‐inflated Poisson regression model with random effects), the current paper derives two conditional Akaike information criteria. The criteria are the unbiased estimators of the conditional Akaike information based on the conditional log‐likelihood and the conditional Akaike information based on the joint log‐likelihood, respectively. The derivation is free from the specific parametric assumptions about the conditional mean of the true data‐generating model and applies to different types of estimation methods. Additionally, the derivation is not based on the asymptotic argument. Simulations show that the proposed criteria have promising estimation accuracy. In addition, it is found that the criterion based on the conditional log‐likelihood demonstrates good model selection performance under different scenarios. Two sets of real data are used to illustrate the proposed method.