AUTOMATIC LAG SELECTION IN COVARIANCE-MATRIX ESTIMATION

We propose a nonparametric method for automatically selecting the numb er of autocovariances to use in computing a heteroskedasticity and aut ocorrelation consistent covariance matrix. For a given kernel for weig hting the autocovariances, we prove that our procedure is asymptotical ly equivalent to one that is optimal under a mean-squared error loss f unction. Monte Carlo simulations suggest that our procedure performs t olerably well, although it does result in size distortions.