Alternative GARCH in Mean Models: An Application to the Korean Stock Market
The purpose of this paper is the theoretical and empirical comparison of alternative GARCH-in-mean models. We examine three GARCH specifications: Bollerslev's (1986) GARCH model, Taylor (1986) - Schwert's (1989) GARCH model, and Nelson's (1991) Exponential GARCH model. In addition, we employ four of the most common forms in which the time-varying variance enters the specification of the mean to determine the risk premium: the quadratic, the linear, the logarithmic and the square root one. For all the aforementioned models we give the auto/cross correlations of the process and its conditional variance. The practical implications of the results are illustrated empirically using daily data on the Korean Stock Price Index (KOSPI).
Authors: | Karanasos, Menelaos ; Kim, J. |
---|---|
Institutions: | Department of Economics and Related Studies, University of York |
Saved in:
freely available
Saved in favorites
Similar items by person
-
Moments of the ARMA-EGARCH Model
Karanasos, Menelaos,
-
Some Exact Formulae for the Constant Correlation and Diagonal M - Garch Models
Karanasos, Menelaos,
-
The Covariance Structure of Mixed ARMA Models
Karanasos, Menelaos,
- More ...