We develop Bayesian methods of analysis for a new class of Threshold Autoregressive models: Endogenous Delay Threshold. We apply our methods to the commonly used sunspot data data set and find strong evidence in favor of the EDTAR model over linear and tranditional threshold autoregressions.

We examine dynamic asymmetries in US unemployment using non-linear time series models and Bayesian methods. We find strong statistical evidence in favour of a two regime threshold autoregressive model. Empirical results indicate that, once we take into account both parameter and model...