Markov-Switching Models with State-Dependent Time-Varying Transition Probabilities

Markov-switching models with covariate-dependent transition functions that are subject to exogenous discrete stochastic changes considered. These associated simultaneous in the covariance structure of observable variables. Simulation experiments carried out assess quality large-sample approximations distributions maximum-likelihood estimator and related statistics such a model, examine implications misspecification due unaccounted breaks mechanism. The practical use model is illustrated by analyzing relationship between Argentinian sovereign bond spreads output growth.