Testing for structural stability in the whole sample

Abstract. Testing for structural stability has attracted a lot of attention in theoretical and applied research. Oftentime the test is based on the supremum of the Wald or Lagrange Multiplier tests when the change is assumed to be in the middle of the sample, that is in the interval [�n] < t < n [�n] for some > 0 and where n denotes the sample size. Recently there has been some work to allow the possibility that the break lies at the end of the sample, i.e. when t 2 n�t; n for some �nite number �t. However, the previous two setups do not include the important intermediate case when t 2 (1; [�n]) [ (n [�n] ; n), or more generally when we do not assume any prior knowledge on the possible location of the break. The aim of the paper is to extend existing results on stability tests in the later scenario for models useful in economics such as nonlinear simultaneous equations and transformation models. Letting the time of the break to be anywhere in the sample might not only be more realistic in applied research, but also it avoids the need to choose either �t or �. In addition we show that contrary to conventional tests such as the CUSUM or the sup[�n]

en

http://eprints.lse.ac.uk/35732/