|GENEVIÉVE GAUTHIER, École des Hautes Études Commerciales, Service de l'enseignement des methodes quantitatives de gestion, 3000, chemin de la Cote-Sainte-Catherine, Montréal, Quebec H3T 2A7, Canada|
|Asymptotic distribution of the EMS option|
Monte Carlo simulation is a commonly used valuation tool in finance. The method is useful for computing prices of derivative securities when an analytical solution does not exist. Recently, a new simulation technique, known as empirical martingale simulation ( EMS), has been proposed by Duan and Simonato (1998) as a way of improving simulation accuracy. EMS has one drawback, however. Because of the dependency among sample paths created by the EMS adjustment, the standard error of the price estimate cannot be obtained by simply using one simulation sample and its asymptotic distribution is unknown. In this paper, we develop a scheme that is capable of overcoming this deficiency. The EMS price estimator is shown to have an asymptotically normal distribution.