Multimodality in GARCH Regression Models.

Journal article

It is shown empirically that mixed autoregressive moving average regression models with generalized autoregressive conditional heteroskedasticity (Reg-ARMA-GARCH models) can have multimodality in the likelihood that is caused by a dummy variable in the conditional mean. Maximum likelihood estimates at the local and global modes are investigated and turn out to be qualitatively different, leading to different model-based forecast intervals. In the simpler GARCH(p,q) regression model, we derive analytical conditions for bimodality of the corresponding likelihood. In that case, the likelihood is symmetrical around a local minimum. We propose a solution to avoid this bimodality.

Authors: Doornik, J; Ooms, M

• Publisher: Elsevier
• Journal: International Journal of Forecasting
• Volume: 24
• Issue: 3
• Pages: 432 - 448
• Publication date: 2008-01-01
• DOI: 10.1016/j.ijforecast.2008.06.002
• Language: English