The Comparisons between Three Option Pricing Models and Black-Scholes Formula in Pricing Performance and Trading Strategy: Evidence from the FTSE 100 Options

• Main Authors: Chia-Ho Han, 韓佳和
• Other Authors: An-Sing Chen
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/02266190038783682458

碩士 === 國立中正大學 === 財務金融所 === 96 === In this paper, I make comparisons between SVSI-J (SV, SVJ), SGT (T, PE, LP), Weibull option pricing models and BS formula in pricing performance and trading strategy. The dataset is FTSE 100 options from 2, January 2002 to 31, December 2004. In the parameters estimated, the least objection function is SVJ model and the worst is LP model. In conclusion, the model allows price jumps to occur or allowing skewness is very important in improving the model fit. In-the-sample trading test, I can draw 3D surface plot for each condition. No matter which models and moneyness, you could make more money when “FTSE 100 Return (t-1) is positive and adopt buy strategy” and “FTSE 100 Return (t-1) is negative and adopt sell strategy”. And you have to be careful when “FTSE 100 Return (t-1) is negative and adopt buy strategy” and “FTSE 100 Return (t-1) is positive and adopt sell strategy”. Because most of the time you could not make money. I also find when FTSE 100 Return (t-1) is negative and adopt “Sell” strategy, pricing errors far from zero are most opportunities let you make money. After that, out-of-the-sample trading test indicates investors can rely on 3D surface plot make more money than without one. In pricing performance test, I adopt distant test to show which model is best to fit “true” statistical density. In distant test, the SVJ model is most to fit the “true” statistical density.