文章基本信息

标题：Band Spectrum Regression for Cointegrated Time Series with Long Memory Innovations
本地全文：下载
作者：D Marinucci
期刊名称：Japanese Studies Programme Papers
出版年度：1998
出版社：Suntory Toyota International Centres for Economics and Related Disciplines
摘要：Band spectrum regression is considered for cointegrated time serieswith long memory innovations. The estimates we advocate are shown tobe consistent when cointegrating relationships among stationaryvariables are investigated, while OLS are inconsistent due to correlationbetween the regressor and the cointegrating residuals; in the presenceof unit roots, these estimates share the same asymptotic distribution asOLS. As a corollary of the main result, we provide a functional centrallimit theorem for quadratic forms in nonstationary fractionally integratedprocesses
关键词：Long-range dependence; band spectrum regression;cointegration