Title: Finite sample performance of generalized covariance estimator
Authors: Joann Jasiak - York University (Canada) [presenting]
Abstract: The finite sample performance of the Generalized Covariance (GCov) estimator is examined for semi-parametric dynamic models with independent identically distributed errors. The GCov estimator is obtained by minimizing a residual-based multivariate portmanteau statistic. It has an interpretable objective function, circumvents the inversion of high-dimensional matrices, and achieves semi-parametric efficiency in one step. We study its finite sample properties in application to the mixed causal-noncausal Vector Autoregressive (VAR) model and examine the effect of the error distribution, the number of autocovariance conditions and the lag.