Discriminating Between Autocorrelation and Misspecification in Regression Analysis: An Alternative Test Strategy
提出一种替代策略，通过有序假设检验和自相关稳健的设定错误检验，区分回归模型中的自相关与模型设定错误，适用于外生回归变量模型，并讨论了滞后因变量模型的处理建议。
A strategy for di scriminating between autocorrelation and misspecification is proposed as an alte rnative to Jerry G. Thursby's(1981) procedure. An ordered sequence of hypothese s is formulated and tested using an autocorrelation robust check for misspecific ation and asymptotic tests of the form of the error autocorrelation model. A dis cussion of responses to the various outcomes of the tests is provided, along wit h comments on the implementation of the procedures. The strategy is only appropr iate for regression equations with exogenous regressors, but suggestions are mad e for analyzing models with lagged dependent variables. Copyright 1987 by MIT Press.