Option Pricing Model Parameter Estimates And Applications

It isn't consistent with the actual financial market that we assume the volatility be a constant in B-S model, and the truth that the "smile" and "skewed" effects of the volatility on exercise price curve are implied in option market data have further proved that volatility is random. So that stochastic volatility (SV) model is proposed in this paper, which is expanded the B-S model, the parameter estimation of the SV model and its application are researched.This paper has elaborated on the SV model, and has mainly researched the parameter estimation of SV model and its application in Shanghai and Shenzhen stock markets. Based on the closing price data of every day from 1995/10/26 to 2005/12/30, the paper use the Efficient Method of Moments (EMM) and Markov chain Monte Carlo (MCMC) method successfully to resolve the question, that we can't directly obtain the likelihood function and the unconditional moment of the SV model for the variance of the average equation is a potential and indirect observed variable. After comparing, two estimated methods have very great accuracy on large sample. Thanks to Winbugs software, MCMC just need establish markov chain to realize posterior distribution, but it requires good operating environment; EMM need choose the proper auxiliary model so that we can calculate its density function and establish moment conditions, but it also require good ability to program. In addition, through the analysis of Chinese stock market based on the SV model, the persistence of volatility is evident; the higher level of average volatility show that there are many noise transaction in Chinese stock market; and the coefficient of the disturbance have a certain distance with the developed countries. Generally speaking, Chinese stock market well be better.