@unpublished{Abadir:2005, author = {Abadir, K and Distaso, W}, title = {Testing joint hypotheses when one of the alternatives is one-sided}, year = {2005} }
TY - UNPB AB - We propose a class of statistics where the direction of one of the alternatives is incorporated.We modify a class of multivariate tests with elliptical confidence regions, not necessarily arisingfrom normal-based distribution theory. The resulting statistics are easy to compute, they do notrequire the re-estimation of models subject to one-sided inequality restrictions, and theirdistributions do not require bounds-based inference. We derive exact explicit distributions, thenprove some desirable properties of our class of modified tests. We then illustrate the relevance ofthe method by applying it to devising an improved test of random walks in autoregressive models withdeterministic components. In this example, the usual alternative to a unit root is one-sided in thedirection of stable roots, while deterministic components are allowed to go either way, and we showthat it is beneficial to take the partially one-sided nature of the alternative into account. AU - Abadir,K AU - Distaso,W PY - 2005/// TI - Testing joint hypotheses when one of the alternatives is one-sided ER -