ANALYSING THE DETERMINANTS OF TERRORISM IN TURKEY USING GEOGRAPHICALLY WEIGHTED REGRESSION

This paper investigates the determinants of provincial terrorism in Turkey taking spatial dimension into account for the time period 1990–2006. Following a traditional global regression analysis, spatial variations in the relationships are examined with geographically weighted regression (GWR) to obtain locally different parameter estimates. Empirical results indicate that increases in income and schooling ratio tend to reduce the provincial average level of terrorism, whereas an increase in unemployment enhances it. Moreover, GWR results indicate that the provincial effects of per capita income and education are more pronounced for the Eastern and South Eastern provinces compared to the Western provinces.     

Is South Africa's 20 years of democracy in crisis? Examining the impact of unrest incidents in local protests in the post-apartheid South Africa

This article assesses South Africa's 20 years of democracy, which has frequently been bedevilled by local protests. This article is particularly interested in those unrest incidents that have been distinct for one reason or another. In South Africa, the surge in local unrest incidents is often linked to police behaviour, which has become increasingly militant and brutal when quelling protests. While the right to protest is constitutional, the damage caused during protest action can be immense owing to acts of violence. Protestors' resorting to violent protest action is often linked to increased frustrations as a result of the government's inability to live up to the expectations of marginalised and poor communities. While this article draws primarily on existing scholarly debates focusing on social protests, the primary objective is to examine the impact of unrest incidents on local protests in post-apartheid South Africa. The article further argues that ‘spaces’ for participation in local governance should be created so that citizens can become architects of development and their own future.     

基于局域模板的STAP非均匀检测器

分析了STAP(Space Time Adaptive Processing)中强干扰目标影响传统非均匀检测器性能的机理,针对强干扰目标引起的非均匀检测器性能下降和传统自适应功率剩余(APR)方法运算量大的问题,提出了一种基于局域模板的非均匀检测器(Local Template NHD,LT-NHD)。该方法不需要进行协方差矩阵求逆的耗时运算,便能获得较高的运算效率;此外,其检测不易受训练样本中的强目标样本影响,能获得比传统方法更为鲁棒的检测性能。实测数据实验验证了本文方法的有效性。     

基于记忆有理函数的功率放大器行为模型

针对功率放大器的非线性特性及记忆效应,提出了一种基于记忆有理函数的功放行为模型。在传统记忆多项式模型和无记忆有理函数模型基础上,构建记忆有理函数模型,并利用共轭梯度法辨识模型系数,同时比较不同记忆深度和不同非线性阶数下的归一化均方误差,获取最佳记忆深度和非线性阶数。采用多载波的WCDMA信号和MRF6S21140H功放来验证模型的有效性,并与记忆多项式模型、无记忆有理函数模型进行了比较。结果表明,记忆有理函数模型在减少系数数目的同时具有更好的逼近精度。     

爆炸冲击作用下加筋板结构变形研究

为了预报加筋板结构在爆炸冲击载荷作用下的变形程度,将加筋板的变形分为整体变形和局部变形,借助数值计算拟合了两者能量分配关系,进而提出了爆炸冲击作用下加筋板结构变形的理论计算方法,并与实验结果进行比较,结果吻合较好。     

基于RSSI线性回归分析的无线传感器网络定位方法

在无线传感器网络（WSN）定位应用中,针对接收信号强度指示（RSSI）容易受到环境的影响不能实现准确测距的问题,提出了一种基于RSSI线性回归分析的定位方法。该方法通过信号衰减模型和线性回归理论相结合,修正实际环境下每个锚节点的测距模型,同时利用相关系数和剩余标准差对测距模型进行评估,制定出更加合理的定位策略。实验表明,采用修正的测距模型和新的定位策略,使得节点定位精度明显提高。     

基于梯度方向二进制模式的空间金字塔模型方法

空间金字塔模型由于其优势在当前图像分类中得到了广泛应用。然而,其码本生成和特征量化这两个环节具有较高的计算复杂度。为了解决这个问题,提出了一种新的局部特征表述———梯度方向二进制模式,首先对图像稠密采样得到多个子图像块,再将每个子图像块均匀划分为2×2个网格,计算每个网格的梯度直方图,然后对所有网格的梯度主方向进行二进制编码并连接为二进制串值,该二进制串值转换的十进制数即为子图像块的特征表述,最后将该特征表述嵌入到SPM模型中。在标准分类数据库上的实验结果证明了本方法在算法耗时和分类精度上均优于基于SIFT的SPM方法。     

军用飞机燃油消耗多元线性回归模型

为了解决军用飞机非标准动作燃油消耗建模问题,利用实际飞参数据,采用多元线性回归方法进行建模分析。建立了燃油消耗量与飞行时间、油箱平均油量、平均高度、平均速度、T1温度、T4温度、N2转速、高度变化量、飞机外挂的多元线性回归模型,并对回归模型进行统计性检验和验证。结论显示:模型拟合度较好,符合军用飞机飞行耗油规律,对军用飞机非标准动作燃油消耗规律研究具有重要意义。     

Sequential learning versus no learning in Bayesian regression models

We consider two regression models: linear and logistic. The dependent variable is observed periodically and in each period a Bayesian formulation is used to generate updated forecasts of the dependent variable as new data is observed. One would expect that including new data in the Bayesian updates results in improved forecasts over not including the new data. Our findings indicate that this is not always true. We show there exists a subset of the independent variable space that we call the “region of no learning.” If the independent variable values for a given period in the future are in this region, then the forecast does not change with any new data. Moreover, if the independent variable values are in a neighborhood of the region of no learning, then there may be little benefit to wait for the new data and update the forecast. We propose a statistical approach to characterize this neighborhood which we call the “region of little learning.” Our results provide insights into the trade‐offs that exist in situations when the decision maker has an incentive to make an early decision based on an early forecast versus waiting to make a later decision based on an updated forecast. © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 532–548, 2014     

基于统计回归模型的突发事件网络舆情研究

突发事件网络舆情传播速度快,在较短的时间内能够引起社会大众的广泛关注,产生较大的影响.通过利用突发事件网络舆情传播规律和酶促反应相似的机理,构建统计回归模型来研究分析突发事件网络舆情的规律,并通过实例进行论证,以期为政府应对突发事件网络舆情提供参考.