基于格兰杰因果关系检验的ISM改进与应用

文章提出了一种基于格兰杰因果关系检验的解释结构模型(ISM)改进方法,该方法能依据客观统计数据识别影响因素的两两关系,进而改善了原有模型中采用专家评价法确定邻接矩阵的主观性困境.通过对物流与经济协同发展水平的影响因素分析,实证检验了解释结构模型改进方法的可行性.     

基于DPN模型的我国出口拉动增加值时滞模型

由于我国加工出口占比较高,学者们提出了区分贸易方式的非竞争型投入产出模型.该模型可以更加准确地测算我国出口对增加值的拉动,但是该模型忽略了出口对增加值拉动的时滞性,这容易产生所有效应都在当年发生的误解.文章基于区分贸易方式的非竞争型投入产出模型构建了各部门出口对增加值拉动的时滞模型.实证结果表明:非加工出口部门的时滞长于加工出口部门的时滞.     

静态金融可计算一般均衡扩展模型构建及货币政策模拟

文章将要素价格完全内生化,并通过引入“货币乘数”和“消费价格指数”,构建了扩展的静态金融可计算一般均衡模型,模拟了货币发行规模变化和存款准备金水平的改变对我国宏观经济、各产业部门和价格水平的影响.结论表明:从货币角度看,价格水平变化并不单一地依赖于货币总量的改变,准备金对货币市场调控效果远强于货币发行,而无论何种宽松的货币政策,央行释放的货币与银行信贷资金增加余额不成比例,银行惜贷现象仍较为严重;从产业角度看,增加货币供给会提高劳动要素报酬而降低资本要素支付,无论采取宽松或是紧缩的货币政策,农业相较于其他产业,对货币的弹性较大.     

基于非平稳面板计量的中国经济动态效率分析

戴蒙德的代际交叠模型从理论上阐述了处于长期均衡增长路径的经济体运行可能是动态无效的.文章基于前人对AMSZ准则的扩展,根据中国1997-2015年的省际面板数据,利用新发展的非平稳面板计量方法进行了实证分析.克服了以往研究指标选取上的争议和小样本带来的估计偏误.结果发现:中国经济运行是动态无效率的,资本存量已经超过了黄金律水平.     

不完全数据多重插补的Bootstrap方差估计

当对插补所得的“完整数据集”使用标准的完全数据统计方法的时候,往往会低估插补估计量的方差.Bootstrap方法(自助法)是非参数统计中的一种重要的统计方法,是基于原始观测数据进行重复抽样,能充分的利用已知数据,不需要对未知总体进行任何的分布假设或增加新的样本信息,进而再利用现有的统计模型对总体的分布特性进行统计推断.本文首先运用多重插补的方法对缺失数据进行了插补,之后利用Bootstrap方法对插补之后的数据进行了插补统计量的方差估计,结果表明运用Bootstrap方法进行插补统计量的方差估计更科学更准确.     

中国十大城市群城镇化水平与城市脆弱性的脱钩关系研究

文章基于中国十大城市群内100个地级以上城市的相关数据,通过构建城镇化水平与城市脆弱性综合评价指标体系,采用因子分析法和Tapio脱钧指数法,实证研究了中国十大城市群及所属城市的城镇化水平与城市脆弱性的脱钩关系,并对其脱钩状态进行判别和分析比较.结果表明:京津翼城市群城镇化水平与城市脆弱性的脱钩状态最理想,辽中南城市群、长三角城市群、山东半岛城市群和川渝城市群均为相对理想的脱钩状态,其他城市群的脱钩状况均不容乐观;在17个中心城市中,仅有洛阳、深圳、成都和西安4个城市的城镇化水平与城市脆弱性的脱钩状态相对乐观,经济发展水平较高的北京、上海和广州3大城市的脱钩状态均不理想.     

Enabling robust assessment of QTc prolongation in early phase clinical trials

Since the implementation of the International Conference on Harmonization (ICH) E14 guideline in 2005, regulators have required a “thorough QTc” (TQT) study for evaluating the effects of investigational drugs on delayed cardiac repolarization as manifested by a prolonged QTc interval. However, TQT studies have increasingly been viewed unfavorably because of their low cost effectiveness. Several researchers have noted that a robust drug concentration‐QTc (conc‐QTc) modeling assessment in early phase development should, in most cases, obviate the need for a subsequent TQT study. In December 2015, ICH released an “E14 Q&As (R3)” document supporting the use of conc‐QTc modeling for regulatory decisions. In this article, we propose a simple improvement of two popular conc‐QTc assessment methods for typical first‐in‐human crossover‐like single ascending dose clinical pharmacology trials. The improvement is achieved, in part, by leveraging routinely encountered (and expected) intrasubject correlation patterns encountered in such trials. A real example involving a single ascending dose and corresponding TQT trial, along with results from a simulation study, illustrate the strong performance of the proposed method. The improved conc‐QTc assessment will further enable highly reliable go/no‐go decisions in early phase clinical development and deliver results that support subsequent TQT study waivers by regulators.     

Schematic saturated orthogonal arrays obtained by using the contractive replacement method

This paper presents a general method of constructing schematic saturated orthogonal arrays of strength two. We consider a class of mixed saturated orthogonal arrays produced from saturated symmetric orthogonal arrays by using the contractive replacement method, study the Hamming distances of the rows, construct their association schemes, and prove that they are schematic. Some examples are given to illustrate this construction method.     

Vine copula constructions of higher-dimensional dependent reliability systems

Copulas have proved to be very successful tools for the flexible modeling of dependence. Bivariate copulas have been deeply researched in recent years, while building higher-dimensional copulas is still recognized to be a difficult task. In this paper, we study the higher-dimensional dependent reliability systems using a type of decomposition called “vine,” by which a multivariate distribution can be decomposed into a cascade of bivariate copulas. Some equations of system reliability for parallel, series, and k-out-of-n systems are obtained and then decomposed based on C-vine and D-vine copulas. Finally, a shutdown system is considered to illustrate the results obtained in the paper.