Efficient and Robust Estimation of GARCH Models

Generalized autoregressive conditional heteroscedastic (GARCH) models have been a powerful tool for modeling volatility. In this paper, we propose an efficient and robust method for estimating the parameters of GARCH models. This method involves a sequence of weights and takes a data-driven weighting scheme to maximize the asymptotic efficiency of the estimators. Under regularity conditions, we establish asymptotic distributions of the proposed estimators for a variety of heavy-or light-tailed error distributions. Simulations endorse our theoretical results. Our approach is applied to analyze the S&P 500 Composite index in the U.S. financial market and run some regression diagnostics to validate the fitted model.