Mixed signals among tests for cointegration
发现，在无协整的原假设下，单方程残差检验与系统检验的p值相关性很低，即使样本量很大也是如此；对132个数据集的实际检验也证实了两种方法p值存在显著差异。
Abstract This paper illustrates that, under the null hypothesis of no cointegration, the correlation of p ‐values from a single‐equation residual‐based test (i.e., ADF or $\widehat{Z}_{\alpha}$ ) with a system‐based test (trace or maximum eigenvalue) is very low even as the sample size gets large. With data‐generating processes under the null or ‘near’ it, the two types of tests can yield virtually any combination of p ‐values regardless of sample size. As a practical matter, we also conduct tests for cointegration on 132 data sets from 34 studies appearing in this Journal and find substantial differences in p ‐values for the same data set. Copyright © 2004 John Wiley & Sons, Ltd.