Munich Personal RePEc Archive

A Range-Based GARCH Model for Forecasting Volatility

Mapa, Dennis S. (2003): A Range-Based GARCH Model for Forecasting Volatility. Published in: The Philippine Review of Economics , Vol. XL, No. 2 (December 2003): pp. 73-90.

[img]
Preview
PDF
MPRA_paper_21323.pdf

Download (159kB) | Preview

Abstract

A new variant of the ARCH class of models for forecasting the conditional variance, to be called the Generalized AutoRegressive Conditional Heteroskedasticity Parkinson Range (GARCH-PARK-R) Model, is proposed. The GARCH-PARK-R model, utilizing the extreme values, is a good alternative to the Realized Volatility that requires a large amount of intra-daily data, which remain relatively costly and are not readily available. The estimates of the GARCH-PARK-R model are derived using the Quasi-Maximum Likelihood Estimation (QMLE). The results suggest that the GARCH-PARK-R model is a good middle ground between intra-daily models, such as the Realized Volatility and inter-daily models, such as the ARCH class. The forecasting performance of the models is evaluated using the daily Philippine Peso-U.S. Dollar exchange rate from January 1997 to December 2003.

UB_LMU-Logo
MPRA is a RePEc service hosted by
the Munich University Library in Germany.