초록 close

In recent years the evidence of predictability has led to a variety of approaches, because the GARCH model cannot capture some important features of the data. The most interesting of these approaches are the asymmetric or leverage volatility models, in which good news and bad news have different impact on volatility. Hence, this paper introduces the models capturing such asymmetric effect, which includes the EGARCH, AGARCH, and GJR models. The asymmetric models found that negative shocks introduced more volatility than positive shocks. The EGARCH model, however, returns unreasonably large conditional variance for large negative shocks. Futhermore, the AGARCH model underestimated the variance for the bad news relative to the GARCH model. This paper, therefore, shows that the GJR model is not only the best at capturing the asymmetric effect, but useful approach to modeling conditional heteroscedasticity.