具有正确收敛性的EM算法在电子政务中的应用

将具有正确收敛性的EM算法应用到电子政务系统中,可最大限度地找到与丢失数据相似的数值,进而保证有效地进行数据挖掘.     

考虑管理弹性的项目成本估算方法

针对项目成本超支问题,考虑项目的实际运作,将管理弹性的价值添加到项目管理中,力求更好地估计项目成本,最终使项目成本得到控制。抓住成本控制中"人"的因素,在给定项目运作方式的基础上,采用反应控制机制,得到更好的项目成本估算,有效地解决项目成本失控问题。     

制造企业主营业务利润质量评价模型研究

制造企业主营业务利润质量是衡量企业整体盈利质量的重要内容,以国资委45家制造企业数据为例,用主成分多元统计分析方法建立制造企业主营业务利润质量评价模型,通过实证归纳得出影响制造企业利润质量最重要的因素是企业对成本的管理和控制能力的结论。     

一类热传导方程的H1-Galerkin混合有限元分析

采用H1 Galerkin混合有限元方法对一类热传导方程的初边值问题,提出了半离散H1 Galerkin混合有限元格式,通过误差分析,得到H1 Galerkin混合有限元解与真解的L2模和H1模的最优阶误差估计.     

The strong uniform convergence of multivariate variable kernel estimates

We show that sup, completely as, where f is a uniformly continuous density on are independent random vectors with common density f, and f_n is the variable kernel estimate Here H_ni is the distance between X_i and its kth nearest neighbour, K is a given density satisfying some regularity conditions, and k is a sequence of integers with the property that log asn     

Checking adequacy of the semiparametric location shift model with censored data

The location shift model is commonly used to quantify the difference between groups in a two-arm study. Nonparametric inference procedures for the location shift parameter with censored observations have recently been extensively studied. However, the validity of these procedures depends heavily on the model assumption. In this article, a class of graphical and numerical methods are proposed for checking the adequacy of the location shift model. Our graphical procedures are much less subjective than the eye-ball method based on the standard Q-Q plot. The proposed methods are illustrated with real-life examples.     

e-SCM系统的收益测算机制设计

作为供应链管理(Supply Chain Management,SCM)的支持手段,e-SCM系统(electronic Supply Chain Management System)的投资滞后成为制约SCM实现的重要障碍。本文借助Preinreich-Luecke定理、相对直接成本以及作业成本法所提供的理论框架,设计了简便易行的e-SCM系统收益测算机制,将其作为e-SCM系统投资决策的先导和收益实现的重要基础。本文的主要结论是:仅当业绩量的成本来源具有可变特质时,e-SCM系统所带来的业绩量才会产生经济收益,以库存为核心的资金占用成本因而成为业绩额的主要来源。仅需提取企业少量运营数据即可进行e化收益测算,客观上增加了测算的可行性。     

The Recursive Kernel Distribution Function Estimator Based on Negatively and Positively Associated Sequences

Let {X _j , j ≥ 1} be a strictly stationary negatively or positively associated sequence of real valued random variables with unknown distribution function F(x). On the basis of the random variables {X _j , j ≥ 1}, we propose a smooth recursive kernel-type estimate of F(x), and study asymptotic bias, quadratic-mean consistency and asymptotic normality of the recursive kernel-type estimator under suitable conditions.