TESTING THE CONSTANCY OF REGRESSION PARAMETERS AGAINST CONTINUOUS STRUCTURAL-CHANGE

A standard explicit or implicit assumption underlying many parameter c onstancy tests in linear models is that there is a single structural b reak in the sample. In this paper that assumption is replaced by a mor e general one stating that the parameters of the model may change cont inuously over time. The pattern of change is parameterized giving rise to a set of parameter constancy tests against a parameterized alterna tive. The power properties of the LM type tests in small samples are c ompared to those of other tests like the CUSUM and Fluctuation Test by simulation and found very satisfactory. An application is considered.