The Robustness of GARCH Option Pricing by the Least-Squares Monte Carlo Simulation

• Main Authors: Nai-cheng Liu, 劉乃誠
• Other Authors: 王之彥
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/17003364152186036086

碩士 === 國立中興大學 === 財務金融學系 === 93 === In this thesis, we study how to price the GARCH option using the Least-Squares Monte Carlo simulation approach, which was introduced by Longstaff and Schwartz (2001). They suggested that it is unsuited to pricing the path-dependent American-style GARCH option. However, it is our finding that Least-Squares simulation framework is applicable to this problem. Furthermore, we also examine the robustness of this technique, where the affects of different basis functions or different number of regressors are examined. The numerical results show that this technique is generally robust, but the degree of robustness still has some connection to basis functions, number of terms, the parameters of the GARCH process and the least-squares-solving method.