A0311
Title: On a measure of lack of fit in nonlinear cointegrating regression
Authors: Ke Zhu - University of Hong Kong (Hong Kong) [presenting]
Abstract: A Portmanteau test is proposed for the adequacy of nonlinear cointegrating regression models. This Portmanteau test is shown to be asymptotically pivotal, and is applicable to a wide class of integrable and non-integrable regression functions with endogenous regressors derived by either short or long memory innovations. Moreover, the applicability scope of this portmanteau test is generalized to include an additive nonlinear cointegrating regression model, whose consistency results are investigated as an independent interest. Finally, the importance of this portmanteau test is demonstrated by simulated and real data.
