Authors: Anibal Emiliano Da Silva (University of Southampton), Jesus Gonzalo (Universidad Carlos III de Madrid), Jean-Yves Pitarakis (University of Southampton),
Paper Number: 1803
Abstract:
We introduce a set of test statistics for assessing the presence of regimes in out of sample forecast errors produced by recursively estimated linear multiple predictive regressions. These predictive regressions can accommodate multiple predictors that are highly persistent with potentially different degrees of persistence. Our method is also designed to be robust to the chosen starting window size so as to avert data mining concerns. Our tests are shown to be consistent and to lead to null distributions that are free of nuisance parameters and hence robust to the degree of persistence of the predictors.