This paper proposes two types of autocorrelation tests based on a methodology that uses an auxiliary regression, named Gauss–Newton regression. All tests are derived considering that the regression function contains contemporary values of endogenous variables, situation in which the model is estimated using the nonlinear instrumental variables method. The first type of test intends to identify the presence of serial correlation, whether genuine or not. The second type of test is proposed to distinguish the genuine serial correlation from the non-genuine serial correlation, being the latter an evidence of misspecification. This study also shows that this second type of test, called the “Common Factor Restrictions” test, can be deduced as a χ2 test or as a t test.
Regression Function Serial Correlation Consistent Estimate Restricted Version Instrumental Variable Method
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in to check access.
Supported by FCT under the Annual Financial Support Program for Research Units granted to CEC and CMUP.
Godfrey, L.G.: Misspecification Tests in Econometrics. Cambridge University Press, Cambridge (1988)CrossRefGoogle Scholar
Hendry, D.F., Mizon, G.E.: Serial correlation as a convenient simplification not a nuisance: a comment on a study of the demand for money by the bank of England. Econ. J. 88, 63 (1978)CrossRefGoogle Scholar
Mizon, G.E., Hendry, D.F.: An empirical and Monte Carlo analysis of tests of dynamic specification. Rev. Econ. Stud. 47, 21–45 (1980)CrossRefzbMATHGoogle Scholar
Rebelo, E., Gonçalves, I.: Testes ao Modelo de Aitken. In: Proceedings of the IX Congress of the Statistics Portuguese Society, pp. 319–326 (2002)Google Scholar
Rebelo, E., Valle, P., Nunes, R.: Nonnested testing for competing autoregressive dynamic models estimated by instrumental variables. Commun. Stat. Theory Methods 41(20), 3799–3812 (2012)CrossRefzbMATHGoogle Scholar