The American Put: Computational Issues and Value Comparisons
比较了四种美式看跌期权定价方法，包括一种解析法和三种数值法，在Black-Scholes框架下分析股息假设和提前行权处理差异对模型价值的影响。
principally to the call contract, the put option has enjoyed a good deal of attention recently. In particular, the early exercise property of American options presents valuation difficulties for puts that are not shared by calls. This obstacle has retarded theoretical progress to some extent. The SEC of 1977 on option market expansion had its heaviest effect on puts, which were only beginning a pilot program at the time. Hence, there was a (coincidental) blunting of practitioner enthusiasm regarding puts. There are indications of progress in both areas. First, a number of approaches have been set forth to deal with the problem of American put valuation. In this paper, I characterize the four principal put valuation competitors in some detail. I confine my attention to a Black-Scholes [ 1] environment with respect to the stochastic process that generates stock returns. One method is analytical in nature, while the other three methods are numerical procedures. Dividend assumptions and the treatment of the possibility of early exercise are not uniform among the valuation procedures. These differences lead to three distinct sets of model put values.' In the next section, I describe the important features of each put valuation method and discuss dividend and early exercise assumptions. The second indication of progress was the relaxation of the moratorium in May 1980. The number of put classes (i.e., stocks with listed put options) expanded from 25 to 167 in the following three months. By August 1981 the number of put classes exceeded 240. This growth will likely continue until all stocks with