基于混合Ⅰ型删失数据的威布尔模型精确推断（英文）

威布尔分布是可靠性和寿命测试试验中常用的模型.本文中,我们考虑了基于混合Ⅰ型删失数据的威布尔模型精确推断.我们得到了威布尔分布未知参数最大似然估计的精确分布以及基于精确分布的置信区间.由于精确分布函数较为复杂,我们也给出了未知参数的另外几种置信区间,比如,基于近似方法的置信区间,Bootstrap置信区间.为了评价本文的方法,我们给出了一些数值模拟的结果.     

Chernoff-Savage统计量的Bootstrap逼近

本文将Efron提出的Bootstrap方法应用于具有广泛应用的Chernoff-Savage统计量，讨论了它的Bootstrap分布的渐进性质，证明了该统计量的规则化形式的Bootstrap分布弱收敛干正态分布．     

基于Bootstrap的厚尾相依序列持久性变点检验

基于修正方差比率函数给出一种检验厚尾序列持久性变点的统计量.在无变点的假设下得到了统计量的渐近分布.为避免检验渐近分布中的厚尾指数,构造Bootstrap抽样方法来确定渐近分布的经验临界值.数值模拟研究结果说明修正方差比率统计量及Bootstrap抽样方法的有效性.     

随机加权法

在讨论置信区间和估计误差的分布等问题时,Efron 提出了 Bootstrap 方法.为说明 Bootstrap 方法,我们用均值估计的误差分布的计算这个例子说明之.设 X_1,…,X_n～iidF,考虑均值μ_1∫xdF 的估计误差     

多总体二阶随机占优对无约束的检验问题

该文将利用保序回归估计和Bootstrap方法对多总体二阶随机占优对无约束的检验问题进行研究,具体步骤如下:首先,利用保序回归估计和经验分布函数构造检验统计量;然后,利用Bootstrap方法给出检验问题的临界值和p值;最后,通过Monte Carlo模拟来说明该文所提出方法的可行性.     

评估系统可靠度置信下限的随机模拟方法

由已知寿命分布和测试数据的五个不同类型部件,组成一串并联系统,本文应用Monte Carlo方法,Bootstrap方法,和Double Monte Carlo方法给出该系统的可靠度置信下限.     

基于联合II型删失数据的两总体指数模型的精确统计推断（英文）

在本文中,我们讨论两指数总体的位置参数和尺度参数的统计推断问题.利用极大似然方法,在联合II型删失数据的情形下给出参数的精确分布以及相关精确统计推断结果.将枢轴量表示为标准指数随机变量的线性函数,并且给出枢轴量的条件精确分布,这个条件精确分布的一个很大优点是计算比较简单.利用条件精确分布,可以获得枢轴量的精确分位数.为了说明本文方法的优劣,我们也提供Bootstrap方法构造参数置信区间的相关结果.最后将理论结果,进行了部分数值模拟实验,这些数值结果列在相应的表格里.     

火工品可靠性试验数据的综合分析方法

升降法试验数据和固定刺激量下的成败型试验数据,是两种最常见的火工品可靠性试验数据．我们应用Markov链,研究了升降法试验数据下,感度分布参数的极大似然估计的特性．在此基础上,应用Bootstrap方法和Bayes方法,给出了综合分析两种试验数据的方法．最后,将该方法应用于520底火的可靠性鉴定,得出了有益的结论．     

独立非同分布情况的光滑Bootstrap逼近

Bootstrap方法一般只适于iid情况,当抽样X_1,…,X_n为独立但不是同分布情况时,本文采用光滑Bootstrap方法,并证明了对于均值这种逼近与iid时Bootstrap逼近具有同样理想的结果。     

右删失情形下平均生存时间的加权Bootstrap推断

在右删失情形下,基于一类合成数据,采用加权Bootstrap方法获得了平均生存时间的加权Bootstrap估计及其加权Bootstrap分布,并就权重是否独立两种情形,证明了此估计的相合性及此分布近似的有效性.基于此,构造了平均生存时间的置信区间.在数值模拟中,取权为Dirichlet(n;1,…,1),并从覆盖概率和区间长度角度,比较了加权Bootstrap和渐近正态逼近产生的置信区间.     

Bootstrap tests for the equality of distributions

Testing equality of two and k distributions has long been an interesting issue in statistical inference. To overcome the sparseness of data points in high-dimensional space and deal with the general cases, we suggest several projection pursuit type statistics. Some results on the limiting distributions of the statistics are obtained. Some properties of Bootstrap approximation are investigated. Furthermore, for computational reasons an approximation for the statistics the based on Number theoretic method is applied. Several simulation experiments are performed.     

Bootstrap in Markov-sequences based on estimates of transition density

We develop a Bootstrap method in Markov-sequences. This method is based on kernel estimates of the transition density of the Markov-sequence. It is shown that the Bootstrap estimate of the variance of a statistic which is a function of means, is consistent. We also show that the Bootstrap distributions of mean-like statistics and von Mises differentiable statistics converge to appropriate normal distributions. A few simulation results are reported to illustrate the discussion.Most of this work was carried out when the author was at the Department of Statistics, Pennsylvania State University.     

重尾分布的尾部指数估计及沪深股市实证分析

重尾分布尾部指数α的估计依赖于样本中所用顺序统计量个数k的选取.本文介绍了估计α时选择k的两类不同的方法:Sum-plot方法和Bootstrap方法,并对Hall提出的Bootstrap方法作了改进,称为M-Bootstrap方法.本文利用上述方法对已知分布进行Monte-Carlo模拟,研究它们的可行性,然后对上海和深圳两市股指数据进行了实证分析.计算结果表明,上海和深圳股指收益率具有重尾性.是右偏态的,右尾厚于左尾.通过几种方法计算的结果比较发现Sum-plot方法和M-Bootstrap方法在估计重尾指数上精确性较高一些,而且不受异常值的影响.     

两个和K个分布函数相等的自助法检验(英文)

两个和Ｋ 个分布函数相等的检验在统计领域中,长期以来一直受到关注．为了克服高维空间中数据点的稀疏性,并且对一般情况进行处理,我们提出了几个投影追踪类型的统计量,并得到了这些统计量的极限分布的一些结果．对一些自助法逼近的性质进行了研究．更进一步地,由于计算的原因,对统计量的逼近,我们采用了数论方法,并进行了一些模拟试验     

基于投影偏度和投影峰度的投影寻踪自助法的正态性检验

本文研究了多元分布的正态性检验问题,用投影寻踪自助法,获得了投影偏度和投影峰度正态性检验统计量,证明了在零假设成立时,所提出的偏度和峰度检验统计量的极限分布为一高斯过程的上界.为计算机模拟计算提供了有力的手段和依据.     

Statistical inference for the Weitzman overlapping coefficient in a family of distributions

A usual problem in applied statistics is the one related to the estimation of the common area under two distributions, which is usually estimated by means of an overlapping coefficient. Particularly, for the Weitzman overlapping coefficient, we found that it is possible to provide a general expression that facilitates making inferences on this coefficient, for many distributions. This expression depends only on two parameters which are actually functions of the parameters of the selected models, among which we can mention the exponential, Weibull, Gumbel, Fréchet and some other distributions that arise under certain transformations of an exponential random variable. The simplicity of our unifying proposal is illustrated considering three well known distributions that have been individually analyzed in statistical literature. To illustrate the performance of the likelihood confidence intervals obtained for this overlapping coefficient, under our proposal, we carried out some simulation studies that yielded adequate coverage frequencies, and just for the sake of comparison we also computed Bootstrap confidence intervals. A real data set is analyzed to exemplify our proposal.     

二维指数信号模型中参数Bootstrap估计的强相合性

本文主要研究了在统计信号处理当中具有广泛应用的二维带白噪声指数信号模型中参数估计的Bootstrap逼近, 借助于回归模型中Bootstrap逼近的构造方法, 给出了二维指数信号模型参数的自助估计, 并证明了自助估计具有强相合性. 最后采用了Monte-Carlo法对所提的方法进行随机模拟, 模拟的结果表明当噪声不服从正态分布时, Bootstrap方法的估计效果优于最小二乘估计.     

On testing for homogeneity of the covariance matrices

Testing equality of covariance matrix ofk populations has long been an interesting issue in statistical inference. To overcome the sparseness of data points in a high-dimensional space and deal with the general cases, we suggest several projection pursuit type statistics. Some results on the limiting distributions of the statistics are obtained. Some properties of Bootstrap approximation are investigated. Furthermore, for computational reasons an approximation which is based on Number theoretic method for the statistics is adopted. Several simulation experiments are performed.     

ON SOME TESTS FOR ELLIPTICAL SYMMETRY

In this paper, some test statistics of Kolmogorov type and Cramer-von Mises type based on projection pursuit technique are proposed for testing the sphericity problem of a high-dimensional distribution. The limiting distributions of the test statistics are derived under the null hypothesis and any fixed alternative. The asymptotic properties of Bootstrap approximation are investigated. Furthermore, for computational reasons, an approximation for the statistics, based on number theoretic method, is suggested.