Assessing Volatility Forecasting Models: Why GARCH Models Take the Lead

by Matei, Marius
Published in Romanian Journal of Economic Forecasting
, 2009, volume 12 issue 4, 42-65

 Requires a PDF viewer such as Xpdf or Adobe Acrobat Reader


The paper provides a critical assessment of the main forecasting techniques and anevaluation of the superiority of the more advanced and complex models. Ultimately, its scope is to offer support for the rationale behind of an idea: GARCH is the most appropriate model to use when one has to evaluate the volatility of the returns of groups of stocks with large amounts (thousands) of observations. The appropriateness of the model is seen through a unidirectional perspective of the quality of volatility forecast provided by GARCH when compared to any other alternative model, without considering any cost component.

Keywords: volatility, GARCH, forecast, correlation, risk, heteroskedasticity
JEL Classification:
C3, C53, D81