The purpose of this thesis is to determine whether or not Straddle Trading, which is the typical method of volatility trading, is a meaningful profit-loss structure in the KOSPI200 market through two simulations of Straddle Trading, the GARCH (1,1) and Implied Volatility methods.
First, the result of the GARCH (1,1) method is as follows:
Verification of data Time Series for the 5-year Sample Period (2 October 1993 ~ 30 September 1998) through the Unit Root Test proved to be non-stationary. After data that is transferred to the yield rate of the stock price index proved stationary, the GARCH (1,1) model is estimated as the result of ARCH LM shows heteroscedasticity. Out of the Sample Period (1 October 1998 ~ 30 September 1999), volatility is forecast by GARCH (1,1) model and substituting the result for the Black-Scholes and Hull-White Option Pricing Model to draw option theoretical prices. The Wald test on the two theoretical prices showed neither of them is an unbiased estimate to the option market price. From the result of the straddle test on the pre-determined trade rules using the option theoretical price, investment by two option pricing model excluding irregular days proved not to be economical after taking transaction cost into account.
Second, the result of Straddle Trading using 10-day, 30-day historical volatility and implied volatility is as follows:
Implied volatility used Trading Volume-Weighted Volatility of the nearest ATM (At-the-Money) Call and Put Option Implied Volatility. Comparisons of implied volatility with 10-day and 30-day historical volatility are used for the Straddle Trading. As a result, in case trade breaks within 7-day maturity with high volatility, Straddle Trading realized some profit. Transaction costs being considered, however, this strategy was also judged not to be economically useful.
