均匀设计抽样的偏差

该文证明了均匀设计抽样点集的偏差为O（n－1（lnn）d），另外也证明了随机化均匀设计点集的偏差有同样的阶．最后进行了随机模拟．     

正交表型均匀LH设计和抽样

本文提出了一种新的设计和抽样方法－正交表型均匀LH设计和抽样,证明了这种抽样空间是OALH抽样空间的优良子集。这种设计和抽样空间中所有样本都与初始设计具有同阶低偏差等一些优良性质。并将它用于数值积分,证明了对有关参数的估计的方差阶低于其他抽样。同时还给出了有关的模拟结果。     

均匀设计抽样及其优良性质

抽样和设计是计算机试验的一个重要研究课题。本文提出了一种新的抽样方法—均匀设计抽样,研究了它的一些基本性质,并将其应用于数值积分近似计算,这种抽样是王元和方开泰(1981)均匀设计思想的一个发展,也是对Latin Hypercube抽样的一个重要改进,通过与Monte Carlo方法,Latin Hypercube抽样(包括OA-Based Latin Hypercube抽样)和均匀设计的比较,表明了这种抽样的优越性,最后还讨论了在一般分布情形下如何应用这种抽样。     

基于计算机试验的均匀设计

本文在计算机试验的基础上，提出了最小相关准则和最小距离离差准则，并将信息论中的Hamming距离和Lee距离引入到计算机试验中，证明了均匀设计在Hamming距离下的最优性和部分好格子点均匀设计在Lee距离下的最优性．基于偏差的考虑，给出了一类新的好格子点均匀设计和一个学习算法，利用这个学习算法，给出了基于Lee距离的最小距离离差准则的均匀设计表的构造方法．通过与已有的好格子点均匀设计和循环拉丁方均匀设计作比较，证明了文中的均匀设计在距离和偏差意义下有更好的均匀性．     

均匀设计抽样的一、二阶矩

（７）中给出并研究了均匀设计抽样（ＵＤＳ）及随机化均匀设计（ＲＵＤ）的一些优良性质，作者给出该设计和抽样的一、二阶矩。     

低偏差OALHD的构造

本文给出了利用均匀设计和正交表构造低偏差OALH设计的方法，该方法构造的设计既有优良的均匀性具有正交设计的均衡性，一个更重要的优点是可以构造较大样本容量的设计点集，本文同时给出了某些参数的均匀设计表，这些设计优于现有的均匀设计，具有实用价值。     

均匀性度量的势函数模型

分析了均匀性度量应满足的基本条件，根据物理学的势和力的模型，提出了均匀性度量的势函数模型，该模型较好地解决了均匀性度量的可计算性、布点的均匀性调整方法等问题；而且具有均匀性度量所应该具有的所有优良性质，如旋转对称性、平移对称性、中心对称性等。最后给出了应用实例，并就低维投影的均匀性问题和势函数模型的改进作了讨论。      

均匀性度量中的密集性偏差与稀疏性偏差

该文根据王元，方开泰［２］的近似偏差（discrepancy）的均匀性准则，定义了理想布点情况下的标准半径，定义了ｍ 维单位子空间Ｃｍ＝［０，１］中两点间的ｆ距离和ｇ距离，由此定义了最大空穴半径和最小空穴半径，提出了均匀性度量的密集性偏差与稀疏性偏差．给出了二维情况 下的计算结果．我们的方法计算量不大，不仅能较好地度量布点的均匀性以及布点在低维投影的均匀性，而且能指导如何调整布点使之尽可能与理想布点接近．     

从设计到抽样

张润楚,王兆军（１９９６）提出了均匀设计抽样,将均匀设计变成抽样．本文给出一种由设计到抽样的一般方法,它可以将任何一个有优良均匀性的设计点集变成所有样本都有同样优良均匀性的抽样。     

任意凸多面体上布点均匀性的度量

均匀性度量是构作均匀设计的基础,本文从距离概念出发,通过对称的方法,得到一种新的距离函数-势函数,并将势函数作为衡量任意凸多面体上布点均匀性好坏的准则.数值例子和多变量Kendall 协和系数检验表明,当试验区域限制在单位立方体上时,势函数与目前常用的两种偏差-中心化L_2-偏差和可卷L_2.偏差在度量布点均匀性方面结论一致.     

Centered -discrepancy of random sampling and Latin hypercube design, and construction of uniform designs

In this paper properties and construction of designs under a centered version of the -discrepancy are analyzed. The theoretic expectation and variance of this discrepancy are derived for random designs and Latin hypercube designs. The expectation and variance of Latin hypercube designs are significantly lower than that of random designs. While in dimension one the unique uniform design is also a set of equidistant points, low-discrepancy designs in higher dimension have to be generated by explicit optimization. Optimization is performed using the threshold accepting heuristic which produces low discrepancy designs compared to theoretic expectation and variance.

     

相对强弱指数的最佳参数组合

本文首先提出了广义均匀设计抽样,并给出了利用广义均匀设计抽样在半连续半离散区域上求取函数最值的方法,之后利用这种方法,针对香港股票市场,给出了技术分析指标－相对强弱指数的最佳参数组合．     

移动平均线的最佳参数组合

本文首先简化了（1）的偏差计算公式，并利用此公式给出某些新的均匀设计表及某些非平衡均匀设计表。其次，提出了广义的均匀设计抽样，最后把随机化均匀设计与广义的均匀设计抽样应用于移动平均线，得到了它的最佳参数组合并得到了改进后移动平均线的最佳参数组合。     

Average equidistribution and statistical independence properties of digital inversive pseudorandom numbers over parts of the period

This article deals with the digital inversive method for generating uniform pseudorandom numbers. Equidistribution and statistical independence properties of the generated pseudorandom number sequences over parts of the period are studied based on the distribution of tuples of successive terms in the sequence. The main result is an upper bound for the average value of the star discrepancy of the corresponding point sets. Additionally, lower bounds for the star discrepancy are established. The method of proof relies on bounds for exponential sums.

     

均匀设计抽样在股市投资决策上的应用

本文首先给出了在区域D(b)=｛（x,y):0≤y≤x≤b｝上的均匀抽样,以及利用均匀设计抽样求取函数在D的最大值的序贯方法,最后把此方法应用于股市的技术分析中,即把此序贯方法与股票的动态分析相结合,给出了在香港股市应用动态分析的最佳参数。     

Compound inversive congruential pseudorandom numbers: an average-case analysis

The present paper deals with the compound (or generalized) inversive congruential method for generating uniform pseudorandom numbers, which has been introduced recently. Equidistribution and statistical independence properties of the generated sequences over parts of the period are studied based on the discrepancy of certain point sets. The main result is an upper bound for the average value of these discrepancies. The method of proof is based on estimates for exponential sums.

     

Uniform supersaturated design and its construction

Supersaturated designs are factorial designs in which the number of main effects is greater than the number of experimental runs. In this paper, a discrete discrepancy is proposed as a measure of uniformity for supersaturated designs, and a lower bound of this discrepancy is obtained as a benchmark of design uniformity. A construction method for uniform supersaturated designs via resolvable balanced incomplete block designs is also presented along with the investigation of properties of the resulting designs. The construction method shows a strong link between these two different kinds of designs     

Average equidistribution properties of compound nonlinear congruential pseudorandom numbers

The present paper deals with the compound nonlinear congruential method for generating uniform pseudorandom numbers, which has been introduced recently. Equidistribution properties of the generated sequences over parts of the period are studied, based on the discrepancy of the corresponding point sets. Upper and lower bounds for the average value of these discrepancies are established, which are essentially best possible. These results show that the average equidistribution behavior of compound nonlinear congruential pseudorandom numbers fits well the equidistribution properties of true random numbers. The method of proof relies heavily on estimates of the average value of incomplete exponential sums.