Basic index and reliability principle of steel structure , and geometric significance are analyzed . the main computation methods , such as two order moment method , separation function and separation item coefficient , linear separation , and central point , check point , monte - carlo are presented . reliability analysis under relative variable , boundary determination ( single and double boundaries ) is analyzed 研究了钢结构可靠度基本原理、可靠指标的几何意义，分析了钢结构可靠度计算的一次二阶矩方法、分离函数和分项系数法、线性分离法，以及中心点法、验算点法、蒙特卡罗( monte - carlo )法，提出了变量相关情况下的结构可靠度分析方法、可靠度界限分析方法(单侧界限与双侧界限) ，基于极值分布的可靠度分析，并对钢结构动力可靠度分析方法进行了初步探讨。
Similar to the famous von - mise condition on the extreme value theory of 1 - max style , the judge condition that absolutely continuous distribution function is in the domain of attraction of p - max style distribution function is given . at the same time , the error inequality between samples and true values is obtained , and almost sure convergence theorems on the extreme value theory of p - max style are also given 类似于l - max型极值理论中著名的von - mise条件，本文给出了绝对连续分布函数f落在p - max型极值分布函数的吸收域中的判断条件，给出了样本与真值的误差不等式，并给出了关于p - max型极值理论的几乎处处收敛定理。
In the term of extreme value theory , especially , extreme distribution - ii , the 4th chapter amends power law regulation , which makes up parametric estimating problem of extreme value method . as these results , a new method to estimate var , based on laplace & extreme value - ii distribution is put forward 论文第四章，通过研究极值ii型分布，从顺序统计量的角度出发，研究极值分布的尾部展开的一些性质，提出了一种估计极值分布参数的方法，完善了极值理论的参数估计问题，结合实际应用提出了laplace极值混合分布和一种估计var的新方法。
So it can avoid risk of model and computer rightly the var of extreme event . this article presents the theory of extreme value and character of tail of distribution and gives the example of var with index of shanghai stock market by evt , then compares the var result of different computation methods and concludes that traditional var method is static state model and var with evt is dynamic conservative model and has the ability of forecasting risk out of sample comparing to historical simulation method 本文系统地阐述了极值理论和极值分布特征，以上证指数为例，将极值理论应用于风险价值的计算，并将应用结果与传统var方法计算的结果进行了比较分析，最后得出结论：传统的var计算模型是静态的模型，应用极值理论计算var的模型是动态的、相对保守的模型；与历史模拟法相比较，极值理论具有超越样本的预测能力。
