Taylor and Francis Ltd.;Philadelphia: Taylor & Francis Group
摘要:
摘要： This article extends the empirical martingale simulation (EMS) method from using a risk-neutral measure to using a dynamic measure for financial derivative pricing. Although the EMS is shown to be capable of obtaining consistent estimate of financial derivative prices in a more efficient way than the standard Monte Carlo simulation procedure, it can proceed only under a risk-neutral framework. In practice, however, it is cumbersome to obtain the explicit expression of a risk-neutral model when dealing with a complex model. To alleviate this difficulty, we compute the financial derivative prices under the dynamic model and impose the martingale property on the simulated sample paths of both the change of measure process and the underlying asset prices under the dynamic P measure. Hence, we call this modification the empirical P-martingale simulation (EPMS). The strong consistency of the EPMS is established and its efficiency is performed by simulation in the GARCH framework. Simulation results shows that EPMS has the similar variance reduction as the EMS method in option pricing if the risk-neutral model can be obtained, and is more efficient than the standard Monte Carlo simulation in most cases. 出版者： Philadelphia: Taylor & Francis Group 出版日期： 2014-01-17 出處： Communications in statistics. Theory and methods, 2014-01, Vol.43 (2), p.328-342 資源來源： Taylor & Francis Journals 版權： Copyright Taylor & Francis Group, LLC 2014 版權： Copyright Taylor and Francis Group, LLC 識別號： ISSN: 0361-0926 識別號： EISSN: 1532-415X 識別號： DOI: 10.1080/03610926.2012.661504
