A0198
Title: A new approach to regime switching autoregressions
Authors: Frederik Krabbe - Aarhus University (Denmark) [presenting]
Leopoldo Catania - Aarhus BBS (Denmark)
Andrew Harvey - University of Cambridge (United Kingdom)
Abstract: A new way is discussed to construct regime-switching autoregressions, making use of a non-Markovian unobserved process. It is shown that, in a special case, the likelihood implied by this new specification is identical to the classical Markov switching autoregression one. The general case leads to more flexible specifications with tractable likelihood functions. The statistical properties of the model are discussed, and conditions for the consistency and asymptotic normality of the maximum likelihood estimator are established. An application to macroeconomic variables shows that the new specification leads to better estimates and predictions.
