MIDAS类模型的估计及其应用研究

根据MIDAS类模型的不同形式及内部结构,推导出了MIDAS类模型的非线性最小二乘估计方法的具体过程,并以此分析了高频月度PMI对低频季度GDP的预测能力及作用路径.实证结果表明:MIDAS模型类模型能够提取更多高频变量PMI的月度数据信息,当滞后阶数变动到18阶时,U-AR(1)-MIDAS模型拟合效果及样本内预测精度表现最优.     

消费函数的季节性协整分析-北京数据的实证研究

本文对于１９８７－１９９６年的季度数据分别原数据和对数数据在３个频率处进行了季节性协整检验，分析了风险水平的特征，得出了季节性调整可能破坏协整性的结论     

时间序列分解预测法及周期因素的探讨

本文运用时间序列分解预测法，分析门诊人次的变动规律，预测了季度门诊人次。本法与移动平均比率预测法相比，预测精度提高３１％。本文还探讨了时间序列分解预测的周期因素等有关问题。     

基于灰色理论的恐怖袭击事件预测与分析

鉴于恐怖袭击事件的破坏性和危害性,对恐怖袭击事件进行分析和预测能够为相关部门组织反恐行动提供有效的理论支持,基于恐怖袭击事件发生的时间和地区因素,分析恐怖袭击事件的时空特性,建立了基于灰色理论的恐怖袭击事件预测模型,并根据恐袭事件季度性的特点,分季度对建立的模型进行了案例分析,通过计算和Matlab仿真证实了该模型的可行性,对反恐措施的制定具有一定的参考和指导意义.     

加权累加生成的GM(1,1)模型及其应用

根据灰色系统理论中的新信息优先原理知新信息对认知的作用大于老信息的作用,而传统的累加生成没有体现原始数据中新信息的重要性.针对这一问题提出了加权累加生成的概念,并对加权累加生成在单调性、灰指数规律、凸性等方面的性质进行了研究,得到加权累加生成序列具有单调递增性,具有较强的指数规律,并具有下凸性,然后建立了基于加权累加生成的GM(1,1)模型.通过具体算例的计算表明,加权累加生成的GM(1,1)模型的模拟和预测精度比传统的GM(1,1)模型模拟和预测精度高,从而说明了该法的有效性.     

An Application of Double Exponential Smoothing Technique to Quality Control

休哈特控制图和ＳＵＳＵＭ控制图技术，在制造业得到广泛应用．Ｊ．Ｓ．Ｈｕｎｔｅｒ提出用指数加权滑动平均（ＥＷＭＡ）作为质量管理工具．然而，众所周知存在趋势项的情况下，用ＥＷＭＡ预测往往不是过低就是过高．本文的目的是表明利用双指数平滑（ＤＥＳ）技术可以减少此缺点．文中给出一个例子，它表明双指数平滑法比单指数平滑法，能更好地监控对于目标的偏离．     

大数据背景下中国季度失业率的预测研究——基于网络搜索数据的分析

目前,中国失业率统计存在一定局限,不利于准确及时地反映劳动市场的就业变动,大数据技术的快速发展为中国失业率统计提供新的发展视角.基于网络搜索数据,文章从5种常用的预测方法中筛选出最优的支持向量机回归模型,对中国季度失业率进行了预测研究.研究表明,基于网络搜索数据预测的失业率能够比官方数据更早地反映失业趋势的变化,预测失业率与修正后的失业率水平接近,能够为政府部门提供中国失业状况的政策预警.     

季节性时间序列建模的“教”与“思”

针对季节性时间序列,分别探讨了四种预测模型.以中国国内生产总值季度数据为研究对象,进行统计建模分析.结果表明,四种模型预测误差稍有差异,但总体上都具有较优的预测精度.最后对2021年我国国内生产总值四个季度数据进行了预测,为我国国内生产总值预测分析提供参考.在综合建模分析的基础上,着重培养学生统计建模思想和解决复杂问题的综合能力和高阶思维以及在真实情境中所激发的品格、价值观和使命感.     

基于相空间重构的一阶加权局域预测法在溶解氧预测中的应用

将混沌理论引入到水质预测中,分析海河溶解氧时间序列,进行相空间重构和计算最大Lyapunov指数.结果表明,海河溶解氧时间序列具有混沌特性,可以对其进行短期预测.在此基础上,应用一阶加权局域预测模型进行了预测,预测结果具有一定的精度,为水质预测研究提供了新的思路.     

基于窗口DEA和Malmquist指数的我国上市航空公司运营效率评价

运用窗口DEA模型计算和比较我国6家上市航空公司2012～2016年的相对运营效率值,进行全要素综合评价,并利用Malmquist指数模型对6家航空公司5年间的效率变化进行动态分析,将Malmquist指数分解为技术效率变动(EC)和技术变动指数(TC)分析上市航空公司效率变化的原因.研究结果表明:6家上市航空公司运营效率参差不齐,采用单一机型低成本运营的春秋航空效率最高且保持稳定,东航相对效率平均值最低且稳定性较差;研究期内航空公司全要素生产率整体呈上升趋势,但受宏观经济环境影响有所波动;各公司技术水平均有进步,技术效率变动不大.     

Grey forecasting method of quarterly hydropower production in China based on a data grouping approach

Grey model GM (1,1) has been widely used in short-term prediction of energy production and consumption due to its advantages in data sets with small numbers of samples. However, the existing GM (1,1) modelling method can merely forecast the general trend of a time series but fails to identify and predicts the seasonal fluctuations. In the research, the authors propose a data grouping approach based grey modelling method DGGM (1,1) to predict quarterly hydropower production in China. Firstly, the proposed method is used to divide an entire quarterly time series into four groups, each of which contains only time series data within the same quarter. Afterwards, by using the new series of four quarters, models are established, each of which includes specific seasonal characteristics. Finally, according to the chronological order, the prediction results of four GM (1,1) models are combined into a complete quarterly time series to reflect seasonal differences. The mean absolute percent errors (MAPEs) of the test set 2011Q1–2015Q4 solved using the DGGM (1,1), traditional GM (1,1), and SARIMA models are 16.2%, 22.1%, and 22.2%, respectively; the results indicated that DGGM (1,1) has better adaptability and offers a higher prediction accuracy. It is predicted that China's hydropower production from 2016 to 2020 is supposed to maintain its seasonal growth with the third and first quarters showing the highest and lowest productions, respectively.     

Retrieving seasonally adjusted quarterly growth rates from annual growth rates that are reported quarterly

Most macroeconomic time series are reported on a quarterly basis. However, in many cases, official statistical agencies report annual (or ‘year-on-year’) growth rates rather than the quarterly data themselves. In this paper, we demonstrate how to recover seasonally adjusted quarterly growth rates from annual growth rates that are reported quarterly. As an illustration, we apply the technique to the highly seasonal UK Retail Sales series. Using only the annual growth rates calculated from the underlying data, we find that our procedure generates estimated seasonally adjusted quarterly growth rates that are almost indistinguishable from the ‘true’ seasonally adjusted quarterly growth rates.     

利用加权对称估计量对季节性时间序列的单位根检验

本文提出对季节性时间序列利用加权对称估计量的单位根检验,导出相应统计量的极限分布。用MonteCarlo方法计算经验百分位数及检验势,并对最小平方估计量,简单对称估计量和加权对称估计量的经验检验势作了比较。     

一类综合评价指数编制路径及案例

综合评价指数是一种广义上的统计指数,在讨论了综合评价指数与一般统计指数不同的基础上,引述了综合评价指数一般编制步骤并详细分析研究了综合评价指数编制过程中的权重确定,提出了基于等权设计思想的加权修正综合评价指数编制方法,最后以甘肃省装备制造产业为例进行了展示.     

面向电子商务服务水平的变权分析评价方法

针对电子商务服务水平评价问题,提出一种基于变权分析方法的评价方法。首先,建立电子商务服务水平评价指标体系;其次,采用实数、区间数、三角模糊数、语言变量和直觉模糊数5种不同类型表示其评价信息,提出各个电子商务服务水平关于正理想解的相对满意度计算方法,据此确定电子商务服务水平,并具体给出其变权分析评价步骤;最后,通过实例说明该方法的有效性和合理性。     

独立随机序列加权和的强收敛性

本文研究独立随机变量序列加权和的强收敛性,利用截尾法和Borel-Cantelli引理,证明了加权系数ank为列阵情形的强收敛性,在一般双下标加权系数的加权部分和的强收敛性,并对Jamison型加权部分和情形证明了其强收敛的充要条件,推广了Chow与Teicher(1971)[3]的相应结果.     

The minimum sum representation as an index of voting power

We propose a new power index based on the minimum sum representation (MSR) of a weighted voting game. The MSR offers a redesign of a voting game, such that voting power as measured by the MSR index becomes proportional to voting weight. The MSR index is a coherent measure of power that is ordinally equivalent to the Banzhaf, Shapley–Shubik and Johnston indices. We provide a characterization for a bicameral meet as a weighted game or a complete game, and show that the MSR index is immune to the bicameral meet paradox. We discuss the computation of the MSR index using a linear integer program and the inverse MSR problem of designing a weighted voting game with a given distribution of power.     

Optimization method to find the voter power index in weighted voting systems

The issue of estimating the voter power in weighted voting systems is considered. We show that the numerical values of the weighted power index previously introduced by the author can be determined by a simple minimization scheme. The efficiency of the method is demonstrated by estimating the influentiality of U.S. Congressmen.     

Nyström method for Cauchy singular integral equations with negative index

In this paper, the authors propose a Nyström method to approximate the solutions of Cauchy singular integral equations with constant coefficients having a negative index. They consider the equations in spaces of continuous functions with weighted uniform norm. They prove the stability and the convergence of the method and show some numerical tests that confirm the error estimates.