检索结果-南通市图书馆

Variable Selection for High-dimensional Cox Model with Error Rate Control

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2024年

作者： HE Baihua SHI Hongwei GUO Xu zou changliang ZHU Lixing International Institute of Finance School of Management University of Science and Technology of China School of Statistics Beijing Normal University School of Statistics and Data Science LPMC KLMDASR and LEBPS Nankai University Center for Statistics and Data Science Beijing Normal University at Zhuhai Department of Mathematics Hong Kong Baptist University

Simultaneously finding active predictors and controlling the false discovery rate(FDR) for high-dimensional survival data is an important but challenging statistical problem. In this paper, the authors propose a novel variable selection procedure with error rate control for the high-dimensional Cox model. By adopting a data-splitting strategy, the authors construct a series of symmetric statistics and then utilize the symmetry property to derive a data-driven threshold to achieve error rate *** authors establish finite-sample and asymptotic FDR control results under some mild *** results as well as a real data application show that the proposed approach successfully controls FDR and is often more powerful than the competing approaches.

关键词：

An adaptive lack of fit test for big data

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Statistical Theory and Related Fields 2017年第1期1卷 59-68页

作者： Yanyan Zhao changliang zou Zhaojun Wang Institute of Statistics and LPMC Nankai UniversityTianjinChina

New technological advancements combined with powerful computer hardware and high-speed network make big data *** massive sample size of big data introduces unique computational challenges on scalability and storage of statistical *** this paper,we focus on the lack of fit test of parametric regression models under the framework of big *** develop a computationally feasible testing approach via integrating the divide-and-conquer algorithm into a powerful nonparametric test *** theory results show that under mild conditions,the asymptotic null distribution of the proposed test is standard ***,the proposed test benefits fromthe use of data-driven bandwidth procedure and thus possesses certain adaptive *** studies show that the proposed method has satisfactory performances,and it is illustrated with an analysis of an airline data.

关键词： Adaptive test asymptotic distribution divide-and-conquer algorithm massive dataset model specification test

Change-points detection in high-dimension

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Change-points detection in high-dimension

第十届海峡两岸统计与概率研讨会

作者： Yunlong Wang changliang zou Zhaojun Wang Institute of Statistics and LPMC Nankai University

Chang-points detection is an integral component of statistical modeling and es-timation. For high dimensional data, classical methods based on the Mahalanobis distance are typically inapplicable. We propose a novel testing procedure that combines a modi ed Euclidean distance and the extreme statistic. The null distribution of our test statistic is asymptotically normal and the test performs well with nite samples. The number of change-points is determined by a new Schwarzs information criterion together with a pre-screening procedure, and the locations of the change-points can be estimated via the dynamic programming algorithm and the use of the intrinsic order structure of the objective function. Under some mild conditions, we show that the new method provides consistent estimation with an almost optimal rate. Simulation studies show that the proposed method has satisfactory performance of identifying multiple change-points in terms of power and estimation accuracy.

关键词： Asymptotic normality Dynamic Programming Feature Screening High-dimen-sional homogeneity test Multiple change-point detection Sparse signals

The AEWMA Control Charts Based on Likelihood Ratio ForPreliminary Analysis

在线全文

cnki会议

学校读者我要写书评

暂无评论

The AEWMA Control Charts Based on Likelihood Ratio ForPrelim...

中国现场统计研究会第12届学术年会

作者： changliang zou,Yi Dai, School of Mathematical Sciences Nankai University, Tianjin 300071, PR China

Two adaptive exponentially weighted moving average (AEWMA) control charts based on likelihood ratio statistics are proposed for preliminary analysis of individual observations. One focuses on detecting the shifts in location in the historical data and the other is more general can detect a shift in either the location or the scale or both. The simulated results show that the proposed two control charts are, respectively, superior to their competitors in the detection of sustained shifts and linear trend shifts.

关键词：

在线全文

cnki会议

学校读者我要写书评

暂无评论

大规模数据分析中基于外推的调节参数选取

中国科学：数学 2022年第6期52卷 689-708页

作者：任好洁邹长亮李润泽上海交通大学数学科学学院上海200240 南开大学统计与数据科学学院天津300071 Department of Statistics The Pennsylvania State UniversityUniversity ParkPA 16802USA

许多统计建模方法需要通过一个或多个调节参数来控制模型的复杂性.这些调节参数可以是非参数回归或密度估计中核光滑方法的带宽,也可以是在高维建模中利用正则化方法进行特征选择的相关正则化参数.调节参数选择在统计建模和机器学习中... 详细信息

许多统计建模方法需要通过一个或多个调节参数来控制模型的复杂性.这些调节参数可以是非参数回归或密度估计中核光滑方法的带宽,也可以是在高维建模中利用正则化方法进行特征选择的相关正则化参数.调节参数选择在统计建模和机器学习中起着关键作用.对于大规模数据分析,诸如基于信息准则的网格点搜索等常用的调节参数选择方法往往需要巨大的计算成本.即使利用并行计算平台,常用方法的可行性仍值得怀疑.本文旨在开发一种快速算法来有效地近似最优调节参数.该算法需要(a)假设一个参数模型来描述最优调节参数与样本大小之间的趋势,(b)利用子采样数据拟合模型并建立趋势,进而(c)将这种趋势外推至大样本的情形.为了确定子采样样本大小,在给定总计算成本预算的约束下,本文推导子采样的理论最优设计,并进一步证明估计的设计具有渐近最优性.本文的数值研究表明,在多个统计应用中,即使利用一个简单的两参数幂函数模型,所提出的算法所挑选的调节参数与基于全数据集所得到的调节参数几乎一致,同时显著地减少了计算时间和存储空间,表现出明显的优势.

关键词：渐近性外推法非线性模型最优设计预测误差正则化方法

A New Kind of Nonparametric Control Chart

在线全文

学校读者我要写书评

暂无评论

A New Kind of Nonparametric Control Chart

中国现场统计研究会第12届学术年会

作者： Chunguang Zhou, changliang zou,Zhaojun Wang School of Mathematical Sciences Nankai University, Tianjin 300071, PR China LPMC of China

In this paper we present a new type control chart for monitoring the mean of the process based on empirical likelihood ratio confidence intervals. The constructions of our proposed chart are completely nonparametric and no distributional assumptions are required. A limited but representative simulation is included and they are very supportive of the proposed control chart.

关键词： Nonparametric control chart Empirical likelihood Bartlett Correction.

A Comparison of Control Charts for Detecting Variance Shift in Phase I

在线全文

cnki会议

学校读者我要写书评

暂无评论

A Comparison of Control Charts for Detecting Variance Shift ...

中国现场统计研究会第12届学术年会

作者： Yi Dai,changliang zou, Zhaojun Wang School of Mathematical Sciences Nankai University, Tianjin 300071, PR China LPMC of China

In this article we will give two alterative control charts for detecting the shift in the variance of the process mean in Phase I. One is a CUSUM chart based on likelihood ratio statistics and the other is a Shewhart chart based on the F statistics which is the ratio of two sample variances, each divided by its population variance. The simulated results show that the proposed two control charts are, respectively, superior to their competitors in the detection of sustained shifts.

关键词：

在线全文

cnki会议

学校读者我要写书评

暂无评论

基于小波的预分析稳健控制图(英文)

应用概率统计 2008年第3期24卷 274-288页

作者：周纯光邹长亮王兆军南开大学数学科学学院统计学系天津300071

本文提出了一种基于小波的预分析控制图,用于监测第一阶段过程控制中的均值飘移.该方法的基本思想是利用多个不同尺度下的一些相对较大的小波系数来探查过程中的异常变化.这个控制图具有很好的稳健性,所以特别适用于像预分析过程中这样... 详细信息

本文提出了一种基于小波的预分析控制图,用于监测第一阶段过程控制中的均值飘移.该方法的基本思想是利用多个不同尺度下的一些相对较大的小波系数来探查过程中的异常变化.这个控制图具有很好的稳健性,所以特别适用于像预分析过程中这样的对分布信息知之甚少的情形.

关键词：非参数小波统计过程控制预分析

在线全文

学校读者我要写书评

暂无评论

我国商业银行社会责任体系的构建

金融发展研究 2011年第6期 60-62页

作者：史常亮邹昊北京林业大学经济管理学院北京100083

随着企业社会责任理论研究和实践运作的推进,商业银行的社会责任问题成为社会的新热点。本文在分析了商业银行社会责任含义的基础上,论述了我国商业银行应当承担的主要社会责任,进而从其承担社会责任的内部动力机制和外部动力机制两个方... 详细信息

随着企业社会责任理论研究和实践运作的推进,商业银行的社会责任问题成为社会的新热点。本文在分析了商业银行社会责任含义的基础上,论述了我国商业银行应当承担的主要社会责任,进而从其承担社会责任的内部动力机制和外部动力机制两个方面,探讨了相应对策。

关键词：商业银行社会责任动力机制