值分布理论

This article estimates the extreme loss distribution with extreme value theory, estimates the shape parameter with HKKP estimator, which is unbiased to small sample data, selects the right threshold by selecting smallest mean square error of the estimated cumulative distribution function and empirical cumulative distribution function and based on the extreme value theory method, this article estimates the extreme loss distribution and the p-percentile operational loss under certain believe range for Chinese commercial banks, then calculates the operational risk supervise capital.

本文利用极值理论对中国商业银行操作损失极端值分布进行估计，针对尾参数估计的来用传统Hill佑计对小样本数据容易产生偏倚的情况，提出了采用Hill佑计的改进―小样本无偏估计的HKKP佑计来估计操作损失的尾参数，针对由于阈值确定不准确导致结果偏差大的情况，采用最小化估计的累计概率分布与经验累计概率分布平均平方误差的方法确定较精确的阈值，估计出给定置信水平下操作风险损失的分位数，从而使得中国商业银行操作风险监管资本的测定成为可能。

Based on the data of concentration of harmful algae in Bohai Bay, using extreme value theory, the peak-over-threshold distribution of the algae concentration was analyzed.

基于渤海湾赤潮藻类细胞密度的连续监测数据，运用极值理论研究了藻类细胞密度的超阈值分布特性。

Combining the principle of generalized Paretodistribution fitting the tail of a distribution with compound extreme value distribution theory, this paper puts forward Poisson-GP compound threshold distribution model, and gives the results estimated by MLE, CMM and PWM respectively.

论文结合广义Pareto分布拟合底分布尾部的原理与复合极值分布理论，构建Pois(来源：A0e1e1eBC论文网www.abclunwen.com)son-GP复合超阈值分布，并给出了极大似然法、复合矩法和概率权矩法的估计结果。

In Chapter 3, using the distribution theory, we discuss the precise largedeviations of heavy-tailed random sums and the integral representation of the tailasymptopics of the maximum on a random time interval of a random walk with heavytailed increments.

在第三章，我们利用分布理论对重尾随机和的精致大偏差以及重尾随机游动随机时的最大值的尾渐近性的积分表示讲行了讨论。

As an appliance, by applying the addition defined in this paper, the uniqueness theorem of meromorphic functions obtained by Yang Lo is generalized which consider its multiple values to multiple valued algebroidal functions.

本文明确地定义了加、乘运算，并证明了运算后的结果仍是代数体函数。为进一步的研究完善了理论基础。关于亚纯函数的唯一性定理有许多非常好的成果仁'1。杨乐创造了一种方法，结合重值更深人的研究了亚纯函数的值分布〔，一'了。

In chapter 3, prediction theory of time series is reviewed, then non-linear autoregressive model of time series-feedforward neural networks is presented, at the same time a new method-SVD is proposed to design the architecture of feedforward neural networks. By using a trained network with a sufficient number of hidden nodes, according to the distribution of singular values of hidden nodes outputs, we can decide the optimum number of hidden nodes, then the reduced network will be retrained.

第三章：先回顾了时间序列的预测理论，提出了时间序列的非线性自回归模型-前馈神经网络模型，并提出了一种设计前馈神经网络的新方法-奇异值分解法，初始网络结构足够大，网络训练后，对网络隐层节点输出所形成的矩阵作奇异值分解，根据奇异值分布情况，决定网络隐层节点的个数，优化后的网络重新开始训练。

In fact, this thesis contains the following results: We generalize a unicity condition of three meromorphic functions given by Jank and Terglane in class A to the case of arbitrary q meromorphic functoins.

在这篇论文里，我们利用值分布的理论来探讨半纯函数的共值与唯一性的问题，本文包含了以下的结果：将Jank与Terglane有关三个A类中的半纯函数唯一性的结果推广到任意q个半纯函数的情形；证明了C。

In 1975, Yang Le and Zhang Guanghou published their result on the value distribution theory .

1975年，杨乐、张广厚发表了关于值分布理论的研究。

In 1925, R. Nevanlinna established two fundamental theorems of meromorphic functions and the study of the value distribution theory of modern times started.

Nevanlinna建立了亚纯函数的两个基本定理，开始了值分布理论的近代研究。

It is found that the histogram of the values of the significant waves from 20-min records of every month, quarter-year, and one year also shows that the Weibull distribution is also the best one among all examined distributions.

因此本文进一步建立其它相关波高统计值与示性波高之关系，其关系一旦建立，即可藉示性波高以推得其它波高统计值，而本研究同时比较其理论与实际值之差异，以了解花莲海域之波高统计特性及以Weibull分布理论计算的可行性，此结果对於往后分析花莲港之波浪现象，与工程设计上可作为参考之依据。

What are your goals and strategies for growth?

你的成长目标和策略是什么？

And unto the angel of the church in Sardis write; These things saith he that hath the seven irits of God, and the seven star I know thy works, that thou hast a name that thou livest, and art dead.

3：1 你要写信给撒狄教会的使者，说，那有神的七灵和七星的，说，我知道你的行为，按名你是活的，其实是死的。