|
Individual investors as well as fund managers are often
exposed under the market risk. Index futures provide an
effective tool to hedge the systemtic risk. This thesis
studies the performance of various hedging techniques for
the open-end mutual fund market in Taiwan. The MSCI Taiwan
index futures traded on SIMEX is selected to be the
hedging equipment. Four hedging models, the OLS,
GARCH(1,1), Error-correction model, and the bi-variate
GARCH(1,1) are used to test if any model leads to better
hedging effectiveness for specific types of funds.
Presumably, mutual funds with different investment strategies
hold a variety of spot position and, consequently, incur various
degrees of market risk. The comovement between return
of a fund and index futures should be unique. Also, each
hedging model has its own assumption regarding the
time-series characteristic of the return series. We therefore
expect a better hedging model for a specific type of
mutual fund. The
studying period begins in January 1997 and ends in February
1998. The hedging performance of index funds
significantly outperforms that of the other types of
funds, due to the close comovement between index funds and index
futures. Results show that the ECM and bi-variate GARCH(1,1)
models are significantly better than the OLS and
GARCH(1,1). Although the empirical analysis does not
totally confirm the early expectation for the best fitted
models for particular types of funds, the difference of hedge
performance between the theoretically and empirically
best fitted models is only 5%. This suggests that the
hedging performance is relatively insensitive to the model
selection.
|