Estimation of a mean in the presence of an extreme observation

The performance of Anscombe, semi-Winsorization and Winsorization (A, S and W) rules for dealing with extreme observations are investigated for observations from N(μ, σ²) and the simple case where it is assumed that at most one observation in the sample may be biased, arising from N(μ + aσ, σ²) and the primary objective is to estimate μ when σ is unknown. Each of these rules is separately treated in terms of the estimated standard deviation, range and interquartile range. A Monte Carlo method is used to evaluate certain expectation integrals that arise in the computations. We give the results for sample sizes n = 6, 8, 10, 12, 14, 16, 20, 30, 40, 50, 60, 80, 100 of determining the constants necessary to give ‘premiums’ of 0.01 and 0.05 for each of the rules. The performance of the rules is measured in terms of ‘protection’. Features of the resulting tables are discussed.     

Protein folding pathways extracted by OFLOOD: Outlier FLOODing method

The Outlier FLOODing method (OFLOOD) is proposed as an efficient conformational sampling method to extract biologically rare events such as protein folding. In OFLOOD, sparse distributions (outliers in the conformational space) were regarded as relevant states for the transitions. Then, the transitions were enhanced through conformational resampling from the outliers. This evidence indicates that the conformational resampling of the sparse distributions might increase chances for promoting the transitions from the outliers to other meta‐stable states, which resembles a conformational flooding from the outliers to the neighboring clusters. OFLOOD consists of (i) detections of outliers from conformational distributions and (ii) conformational resampling from the outliers by molecular dynamics (MD) simulations. Cycles of (i) and (ii) are simply repeated. As demonstrations, OFLOOD was applied to folding of Chignolin and HP35. In both cases, OFLOOD automatically extracted folding pathways from unfolded structures with ns‐order computational costs, although µs‐order canonical MD failed to extract them. © 2014 Wiley Periodicals, Inc.     

WSNS中基于Fusion-Bayes的离群点检测

随着无线传感器网络技术的发展,数据采集量越来越大,维数也不断提高。然而现有的离群点检测算法多是面向单维或低维度数据,对此文中提出了基于Fusion-Bayes的离群点检测算法。该检测方法首先利用数据转换技术将不同数据属性转换成统一格式,使得各属性可以进行融合运算;然后再利用贝叶斯方法对融合后的属性进行离群点检测。通过实验得出,多维数据属性融合后的检测结果相比于单维属性或低维属性的检测更加准确、效果更好。     

一种改进的相邻概率随机抽样一致性算法

为了解决图像配准中因外点引起迭代次数激增而导致的配准效率低下的问题,提出一种改进的相邻概率随机抽样一致性算法。首先通过自适应优化内点被采用概率,淘汰外点从而形成正循环采样,结合模型评价函数估计出临时模型并得到临时内点集,再根据内点相邻原则筛选部分内点进行重采样,最后采用预检验技术得到最佳模型参数。实验结果表明：在数据外点率较高的情况下该算法能有效减少迭代次数,缩短算法时间,从而提高了算法效率。     

基于联合稀疏功率谱恢复的机载雷达稳健STAP算法研究

机载雷达信号的空时自适应处理（STAP,Space-Time Adaptive Processing）需要利用样本数据来估计杂波协方差矩阵.非均匀杂波环境中的离群点会使协方差矩阵的估计出现偏差,从而导致信号相消现象.针对此问题,本文提出一种基于联合稀疏功率谱恢复的STAP算法（JSR-STAP）处理非均匀杂波,以克服离群点对正侧视模式机载雷达的STAP性能的影响.JSR-STAP算法在高分辨稀疏恢复的条件下,利用多快拍间杂波和离群点的分布规律和相关性不同,通过范数选择合适的稀疏恢复系数来实现离群点的抑制.Monte Carlo实验证明JSR-STAP算法的稳健性和动目标检测性能均优于传统的STAP算法.     

Comparisons of outlier detection procedures in weibull or extreme-value distribution

We carried out an empirical performance study of outlier detection procedures for Weibull or extreme-value distributions using a mixture model in which a known number of randomly chosen observations are contaminated. Procedures studied were: L(L') based on leaps (differences of adjacent observations divided by expectation), V, Q and W (Mann, 1982), R₁(R'₁), R₂(R'₂), R₃(R'₃) (Dixon, 1950) and G(G') (Grubbs, 1950). Percentage points for statistics L(L'), R₁(R₁), R₂(R'₂), R₃(R'₃) and G(G') were computed empirically for the extreme-value distribution and are tabulated. The procedures L(L') (or equivalently in power V) performed best, with few exceptions, for the contaminated model tested. The Grubb statistic G' performed well in testing for lower outliers. Mann's W , which was best for the labeled slippage was substantially poorer than the others for the mixture model. Dixon's R₁ (R₁) is recommended as a generally useful test for sample sizes in the range investigated (n=5,20)     

基于声信号处理的交通事故自动检测算法

建立交通事故应急救援系统具有很强的现实意义,关键技术是对交通事故进行自动检测。为了保证检测的实时性与准确性,提出一种基于声信号处理的方法,采集车辆周围的声音并进行预处理,使用Harr小波变换提取声信号的频域特征,采用单类支持向量机进行异常点检测,实现了分类判别。按上述方法对交通事故发生时的碰撞信号与正常行驶时的非碰撞信号做了分析,准确的识别出交通事故。仿真实验结果表明与常用的线性判别分析方法相比准确率有了显著提高,而且计算复杂度低,易于在DSP系统上实现,算法的判别性能达到了实用化的程度。     

基于簇的孤立点检测

提出一个新的概念——基于簇的孤立点概念,这个概念的提出有助于理解局部数据的行为,同时也描述了如何利用LDBSCAN算法发现基于簇的孤立点,并为每一个对象计算局部偏离因子.     

高维空间中离群点检测算法研究

提出一种基于改进粒子群优化算法的离群点检测算法,解决高维环境下离群点挖掘效率偏低的问题。新算法能够充分发挥粒子群优化算法全局搜索的优势,并具有k均值算法快速收敛的特点,可避免粒子群优化算法的早熟,减小确定k均值算法聚类中心的计算量等问题。实验表明,该算法在高维环境下可快速有效的挖掘出离群数据的离群支持度,有较好的挖掘效率、准确率和实用性。     

A class of estimable contrasts in an age-period-cohort model

A class of estimable contrasts is defined for cohort effects in an age-period-cohort model. It is useful for detecting a systematic change in cohort effects without suffering from a short term deviation. This together with the follow-up analysis of residuals will give a good insight into the data. Numerical examples are given to illustrate how the method applies.