RT Journal Article
T1 Distribution free specification tests for dynamic linear models
A1 Delgado, Miguel A.
A1 Hidalgo, Javier
A1 Velasco, Carlos
AB This article proposes goodness-of- t tests for dynamic regression models, where regressors are allowed to be only weakly exogenous and arbitrarily correlated with past shocks. The null hypothesis is stated in terms of the lack of serial correlation of the errors of the model. The tests are based on a linear transformation of a Barlett`s Tp-process of the residuals. This transformation approximates the martingale component of the process so that it convergesweakly to the standard Brownian motion under the null hypothesis. One feature of our setup is that we do not require to specify the dynamic structure of the regressors. Due to this, the transformation employs a semiparametric correction that does not restrict the class of local alternatives that our tests can detect, in contrast with other works using smoothing techniques. A Monte-Carlo study illustrates the  fi nite sample performance of the tests.
PB University of Cambridge,  Blackwell
SN 1368-4221
YR 2009
FD 2009
LK https://hdl.handle.net/10016/3743
UL https://hdl.handle.net/10016/3743
LA eng
DS e-Archivo
RD 29 jun. 2024