利用现有定价理论 ,大多数美式衍生证券价格不能通过解析方法确定 ,因而必须通过数值分析方法来估计。目前 ,用于美式证券价格估计的数值分析方法主要有三类 ,即格点分析法、有限差分法和蒙特卡罗模拟 ,另外还有一些其它的有效方法。本文主要就这些方法在近些年来的应用与发展进行分析评述。全文共分五部分 ,第一部分是格点分析法 ;第二部分为有限差分方法 ;第三部分是蒙特卡罗模拟 ;第四部分为其他方法 ;最后是结论与展望。 Using the existing pricing theory, the price of most American-derived securities can not be determined by analytical methods and must therefore be estimated by numerical analysis. At present, there are mainly three kinds of numerical analysis methods used to estimate the price of American securities, that is, lattice point analysis, finite difference method and Monte Carlo simulation. In addition, there are some other effective methods. This article mainly analyzes and comments on the application and development of these methods in recent years. The full text is divided into five parts, the first part is the grid analysis method; the second part is the finite difference method; the third part is the Monte Carlo simulation; the fourth part is the other methods; finally, the conclusion and the prospect.