Burr-ⅩⅡ部件可靠性指标的贝叶斯估计

基于逐步增加Ⅱ型截尾样本,讨论了Burr-ⅩⅡ分布的Bayes估计及其性质.在不同的先验分布及刻度平方误差损失函数下给出了参数,可靠度函数以及失效率函数的Bayes估计,并证明了估计的容许性.最后运用Monte Carlo方法对各结果的MSE进行了模拟比较.结果表明,本文给出的估计结果精度高.     

含位置参数的广义Weibull分布及其置信限估计

通过引入位置参数,建立了能够同时描述浴盆曲线失效率和最大安全寿命的广义Weibull分布模型族,进而采用秩分布理论给出确定该寿命分布族置信限估计的参数化方法。具体分析了目前广泛应用的一种广义Weibull分布,导出该模型参数的置信区间估计与分布函数的置信上、下限曲线函数关系式。文中还进行了模拟验证与实例对比分析。     

两参数Pareto分布逐步首失效样本的Bayes估计

基于新的失效截尾模式——逐步增加首失效截尾, 研究两参数Pareto分布可靠性指标的Bayes估计问题. 在对称和非对称损失下, 获得了可靠性指标的Bayes估计. 对超参数未知情形, 利用一种新方法给出超参数估计. 为了研究估计结果的精确性, 利用Monte-Carlo方法给出一个数值模拟例子.     

两参数Weibull分布环境因子的Bayes估计

基于天地试验环境下Weibull分布形状参数不变的条件,运用Bayes方法对两参数Weibull分布环境因子进行分析,以尺度参数后验密度的商密度作为环境因子的后验密度,给出了在形状参数已知和未知两种情况下,环境因子在平方损失下的Bayes估计。最后运用仿真实例与极大似然估计进行比较,结果表明Bayes估计优于极大似然估计。     

威布尔分布无失效数据的Bayes可靠性分析

在Weibull分布场合下,通过对分布函数进行变换,利用凹凸性方法获得了各检测时刻失效概率之间的关系,进而从实际应用的角度出发,利用先验信息得到了各时刻失效概率的Bayes估计.通过建立参数回归模型,获得了产品可靠性指标的估计.最后通过对实例进行计算,验证了方法的稳定性.     

双边定数截尾情形下指数-威布尔分布参数的Bayes估计

在双边定数截尾情形下．给出两参数指数-威布尔（EW）分布形状参数的Bayes估计。基于无信息先验和共轭先验，并分别考虑均方损失和LINEX损失，给出不同的估计结果。最后运用随机模拟的方法对各种估计结果的优良性进行分析比较。     

外汇风险的极值相关模型

研究在已知某一分量极值出现的情况下,通过多维随机向量的联合条件分布的近似分布,利用半参数方法,给出一个极值相关模型,估计多维随机向量的任意尾部事件概率.应用该方法估计加元和日元资产组合一天的99.9%,99.99%,99.999%VaR值分别为0.0321,0.0439,0.0598.     

Weibull分布环境因子的统计分析

利用有关统计量近似服从正态分布的性质 ,证明了Weibull分布场合环境因子基于参数的最好线性无偏估计的点估计是有偏估计。导出了基于定数截尾样本Weibull分布环境因子的近似无偏估计和近似置信区间。利用模拟方法研究了所给的近似无偏估计和近似区间估计的精度。模拟结果显示 ,所给的近似无偏估计和近似区间估计的精度是相当好的     

基于分布估计算法的备件优化配置

分布估计算法是一类新的进化算法,它通过统计在当前群体中优选出的个体信息给出下一代个体分布的概率估计,用随机取样的方法生成下一代群体.将分布估计算法应用于备件优化问题,提出了基于分布估计算法的武器装备系统备件优化模型,分析了个体选择比例、精英保留等策略对算法的影响.仿真结果表明了文中所采用的分布估计算法在求解备件优化问题时是可靠有效的.     

基于g-h分布度量银行操作风险

根据银行操作风险厚尾性的特点, 采用具有厚尾特点的 g-h 分布度量了银行的操作风险. 在实际运用中根据 g-h 分布定义, 修正了 Tukey 分位数估计; 依据操作风险的高频低危和低频高危的特性, 提出了损失区间法确定损失次数参数. 在 g-h 分布特性的基础上, 得出了 g-h 分布的随机和的在险值. 利用我国银行公开披露的操作风险损失数据进行了实证分析, 结果表明提出的估计法能较好地拟合 g-h 分布的参数. 研究结果对于我国银行定量管理操作风险具有启示作用.     

ADMISSIBILITY OF SIMULTANEOUS ESTIMATION OF SEVERAL LOSSES

In this paper,admissibility and inadmissibility results of simultaneous estimation oflosses of multivariate normal mean estimators are given.Admissible estimators for losses ofusual estimators of multivariate gamma parameters,multivariate binomial parameters andPoisson parameters are obtained respectively.     

STRONGLY CONSISTENT ESTIMATION FOR A MULTIVARIATE LINEAR RELATIONSHIP MODEL WITH ESTIMATED COVARIANCES MATRIX

1 IntroductionSuppose that the true values Of a set of variables satisfy exact linear relationships, but theyare unobserved. We can Obtain the Observed values which are the sums of the true values anderrors of measurement. The model is described as follows. A set of rmdimensional vectors {xh}satisfyB'xk or = 0, k = 1,2,'' t (1.1)where B is an m x p parameter mains with rank p < m and o is a p x 1 parameter vector. Weobserved {Zk}, which are m x 1 random vectors and satisfyZk = ac ed, …     

数论网格法在极大似然估计中的应用

复杂似然函数的多峰性使得极大似然估计的求解存在很大困难。针对这一问题,提出了求解极大似然估计问题的数论网格法。讨论了数论网格法的特点,理论分析了其算法精度,给出了基于数论网格的序贯优化算法的计算步骤,研究了初始搜索区域和算法初始参数的确定方法。最后,以两参数威布尔分布参数极大似然估计为例,给出了极大似然估计的计算过程,比较了序贯优化算法和对分法的估计结果,说明了序贯优化算法的有效性和计算效率。     

逐步增加I型混合截尾试验下B-S部件的可靠性分析

在逐步增加I型混合截尾试验下,利用最大期望(expectation maximization, EM)算法讨论了Birnbaum-Saunders(B-S)部件的参数及可靠度的估计问题。将B-S分布复杂的分布形式转换成较为简单的逆高斯分布和逆高斯分布倒数的等加权和形式,推导出逐步增加I型混合截尾中隐藏变量的后验密度,进而获得部件寿命参数和可靠度的极大似然估计和渐近置信区间。蒙特卡罗数值算例表明,简化后的B-S分布在逐步增加I型混合截尾试验方案中,EM算法经过较少迭代即达到收敛。同时,通过对数据情形I和情形Ⅱ下的估计结果分析,给出了较一般截尾试验方式更为丰富的结论,使得相应的统计结果更加全面和精确。     

Statistical analysis of generalized exponential distribution under progressive censoring with binomial removals

The estimation of generalized exponential distribution based on progressive censoring with binomial removals is presented,where the number of units removed at each failure time follows a binomial distribution.Maximum likelihood estimators of the parameters and their confidence intervals are derived.The expected time required to complete the life test under this censoring scheme is investigated.Finally,the numerical examples are given to illustrate some theoretical results by means of Monte-Carlo simulation.     

MODELLING RANDOM CONVEX SETS

A model is developed for convex-set valued data,which are the Minkowski sum of aconvex parameter set and a convex noise set.This model is generalized to include location and scaleparameters.A further generalization results in a model for convex sets which is analogous to aone-way analysis of variance models.In each case,estimators of the parameters are given which,under certain conditions,are shown to be consistent.Data generated from the model and theestimators are studied.An outlier detection method is examined.     

Approximate conditional likelihood for generalized linear models with general missing data mechanism

The generalized linear model is an indispensable tool for analyzing non-Gaussian response data, with both canonical and non-canonical link functions comprehensively used. When missing values are present, many existing methods in the literature heavily depend on an unverifiable assumption of the missing data mechanism, and they fail when the assumption is violated. This paper proposes a missing data mechanism that is as generally applicable as possible, which includes both ignorable and nonignorable missing data cases, as well as both scenarios of missing values in response and covariate. Under this general missing data mechanism, the authors adopt an approximate conditional likelihood method to estimate unknown parameters. The authors rigorously establish the regularity conditions under which the unknown parameters are identifiable under the approximate conditional likelihood approach. For parameters that are identifiable, the authors prove the asymptotic normality of the estimators obtained by maximizing the approximate conditional likelihood. Some simulation studies are conducted to evaluate finite sample performance of the proposed estimators as well as estimators from some existing methods. Finally, the authors present a biomarker analysis in prostate cancer study to illustrate the proposed method.     

ON ASYMPTOTICALLY MEDIAL UNBIASED ESTIMATORS IN TRVNCATED－TYPE DISTRIBUTION FAMILIES

ＯＮＡＳＹＭＰＴＯＴＩＣＡＬＬＹＭＥＤＩＡＬＵＮＢＩＡＳＥＤＥＳＴＩＭＡＴＯＲＳＩＮＴＲＶＮＣＡＴＥＤ－ＴＹＰＥＤＩＳＴＲＩＢＵＴＩＯＮＦＡＭＩＬＩＥＳＬＩＡＮＧＨｕａ（ＩｎｓｔｉｔｕｔｅｏｆＳｙｓｔｅｍｓＳｃｉｅｎｃｅ，ＡｃａｄｅｍｉａＳｉｎｉｃａ...     

An Additive Hazards Model for Clustered Recurrent Gap Times

In this article, clustered recurrent gap time is investigated. A marginal additive hazards model is proposed without specifying the association of the individuals within the same cluster. The relationship among the gap times for the same individual is also left unspecified. An estimating equation-based inference procedure is developed for the model parameters, and the asymptotic properties of the resulting estimators are established. In addition, a lack-of-fit test is presented to assess the adequacy of the model. The finite sample behavior of the proposed estimators is evaluated through simulation studies, and an application to a clinic study on chronic granulomatous disease (CGD) is illustrated.