作者: David A Dickey , Sastry G Pantula
DOI: 10.1198/073500102753410363
关键词:
摘要: One way of handling nonstationarity in time series is to compute first differences and fit a model the differenced unless also looks nonstationary. In that case, second- or higher-order differencing done. To decide if current degree sufficient, one can look at autocorrelation function for slow decay. A formal statistical test need difference further available willing assume most more will render stationary. this article, we present proper sequence tests allows practitioner handle cases which high order may be needed. The not traditional sequence, begins with single unit root.