文章基本信息

标题：Edgeworth Expansions for Spectral Density Estimates and Studentized Sample Mean - (Now published in 'Economic Theory', 17 (2001), pp.497-539
本地全文：下载
作者：Peter M Robinson ; Carlos Velasco
期刊名称：Japanese Studies Programme Papers
出版年度：2000
卷号：1
出版社：Suntory Toyota International Centres for Economics and Related Disciplines
摘要：We establish valid Edgeworth expansions for the distribution of smoothednonparametric spectral estimates, and of studentized versions of linearstatistics such as the same mean, where the studentization employs such anonparametric spectral estimate. Particular attention is paid to the spectralestimate at zero frequency and, correspondingly, the studentized samplemean, to reflect econometric interest in autocorrelation-consistent or long-runvariance estimation. Our main focus is on stationary Gaussian series, thoughwe discuss relaxation of the Gaussianity assumption. Only smoothnessconditions on the spectral density that are local to the frequency of interestare imposed. We deduce empirical expansions from our Edgeworthexpansions designed to improve on the normal approximation in practice, andalso a feasible rule of bandwidth choice
关键词：Edgeworth expansions; nonparametric spectral estimates;stationary Gaussian series; studentized sample mean; bandwidth choice