Are Implied Volatilities Consistent with the Stochastic Properties of Underlying Asset Return? An Empirical Analyze on S&P 500 Index Options Market

碩士 === 國立成功大學 === 財務金融研究所 === 94 === This study is aimed to answer this question:” Are implied volatilities consistent with the stochastic properties of underlying asset return?” The importance of this study is that we make the comparison between actual implied volatilities and underlying return vol...

Full description

Bibliographic Details
Main Authors: Cheng-Hsun Lin, 林政勳
Other Authors: Tse-Shih Wang
Format: Others
Language:en_US
Published: 2006
Online Access:http://ndltd.ncl.edu.tw/handle/06016705989473047422
Description
Summary:碩士 === 國立成功大學 === 財務金融研究所 === 94 === This study is aimed to answer this question:” Are implied volatilities consistent with the stochastic properties of underlying asset return?” The importance of this study is that we make the comparison between actual implied volatilities and underlying return volatility with certain specifications that few studies discuss. Though Heynen, Kemma and Vorst(1994) have the similar study, we develop term structure model in the risk-neutral world, rather than Heynen’s model in the real world. Moreover, unlike other studies, volatility risk premium is considered in our term structure model which is developed from the average of expected volatility assumption. Three major findings are as follows. First, all models may not be well descriptive about implied volatility behavior, which suggests the average expected volatility assumption does not hold. Second, the estimated mean reversion parameter of volatility in the term structure model is larger. A larger mean reversion parameter suggests a flatter term structure curve and more possibility that spot volatility reverts to the long-term mean variance level. Third, there should be theoretical equivalent relation on the mean reverting parameter between any two models among three models in this study. However, the empirical result does not support the equivalent relation among models.