THURSDAY, February 21^st, 2008

Speaker:       Jonathan Hill (Economics, Univ. North Carolina)

Title:            Tail Trimmed Sums for Dependent, Heterogeneous Data, with Applications to Robust GMM

Abstract:       We establish sufficient conditions for asymptotic normality of an intermediate tail-trimmed sum for dependent, heterogeneous data with arbitrarily thin or thick tails. The sum is self-standardized with a kernel variance estimator of the trimmed sum. The resulting central limit theory applies to linear and nonlinear distributed lags, FIGARCH and stochastic volatility processes, all with short or long memory and thin or thick-tailed shocks, including IGARCH and explosive GARCH. The theory is applied to a robust M-estimator of GARCH data when the data may be heavy-tailed, yet standard asymptotics are desired.