The Empirical Analysis of Volatility Smile in Taiwan Options Market

• Main Authors: Jia-Syun Li, 李佳勳
• Other Authors: Jen-Tsung Huang
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/73957138142364527309

碩士 === 國立中山大學 === 財務管理學系研究所 === 101 === In this study, we use the regression by Pena (1999) as theoretical implied volatility estimation, and assuming three variables : “The number of days”, ”Volume” and “Spread” as main variables. The real implied volatility as a dependent variable regression model is estimated by the Black model. We use the daily data of Taiwan Stock Index Options as trading objectives. Finally we use the theoretical implied volatility theoretical to get the theoretical price and take buy low and sell high price strategy to observe the average return changes of three different variables value. The result of call, we find that the number of days has negative relationship with average return. And the period of monthly data (21 days) has the best average returns; Volume has a positive relationship with average return, but more than the threshold 1500 is not; Spreads is not significant relationship. After consideration of dynamic hedging, the number of days still remain a negative relationship with average return; Volume sensitivity direction is reversed; Spreads sensitivity and significance are increasing. The result of put, the spread and he number of days conclusion similar to the call analysis both have negative relationship with average return, but do not present a significant impact; Volume is only significant variable with average returns but more than the threshold 1500 the impact is not significant. After dynamic hedging, all results similar the call conclusion, but hedge profitability is poor, making the average return after hedging decline. Most winning rate less than 50%, is that strategies for capturing market volatility in poor market condition is not well, resulting in low winning rate.