Hedgers should estimate an optimal hedge ratio to manage or reduce the risk of the portfolio and the estimation of hedge ratio depends on the models used. Owing to the nonstationary for the time series, the hedge ratio estimated by the Classic Regression Model might be quite in possibility of wrong. In case of using the hedge ratio estimated incorrectly, it incurs a result like increasing hedging cost rather than reducing market risks of spot portfolios. An Error Correction Model can improve many problems incurred by the nonstationary for time series and that’s why it widely used.
Targeting of currents and futures index for KOSPI 200, this paper will show an estimated hedge ratio based on both the Error Correction Model and the Classic Regression Model and analyze the comparison of a power of explanation and a power of forecast for each model. The hedge ratios which were calculated by two models will also apply to the dynamic hedging strategy and analyze results from portfolio insurance strategies. Finally, I will analyze empirically whether this portfolio insurance strategy will be usefully applied to the Korean market through all the process that I mentioned above.
From the result of the empirical analysis, it appears there is cointegration between KOSPI 200 current index and KOSPI200 futures index, the hedge ratio estimated by the Error Correction Model is different from one estimated by the Classic Regression Model and when it comes to a power of explanation and a power of forecast, the Error Correction Model is superior to the Classic Regression Model.
Conducting dynamic hedging with these hedge ratios, it results as followings. First, when KOSPI200 indices show a downward trend during the whole analyzed period, using the dynamic hedging strategy can decrease losses for the indices` fall by effectively conducting insurance strategies. Second, the result for the dynamic hedging with a hedge ratio estimated by the Error Correction Model was superior to the one with a hedge ratio estimated by the Classic Regression Model. Third, from the comparison for everyday readjustment strategy considering tracking errors, an adjustment strategy exceeding 3% over a market move discipline and an adjustment strategy exceeding 5% over a market move discipline, using the everyday readjustment strategy was most superior in the aspect of the fund value but taking an adjustment strategy exceeding 5% over a market move discipline was more advantageous.
