One goal of this article is to develop an efficient Metropolis–Hastings (MH) algorithm for estimating an ARMA model with a regime-switching mean, by designing a new efficient proposal distribution for the regime-indicator variable. Unlike the existing algorithm, our algorithm can achieve reasonably fast convergence to the posterior distribution even when the latent regime-indicator variable is highly persistent or when there exist absorbing states. Another goal is to appropriately investigate the dynamics of the latent ex-ante real interest rate (EARR) in the presence of structural breaks, by employing the econometric tool developed. We show that excluding the theory-implied moving-average terms may understate the persistence of the observed EPRR dynamics. Our empirical results suggest that, even though we rule out the possibility of a unit root in the EARR, it may be more persistent and volatile than has been documented in some of the literature.

While Perron and Wada (2009) maximum likelihood estimation approach suggests that postwar US real GDP follows a trend stationary process (TSP), our Bayesian approach based on the same model and the same sample suggests that it follows a difference stationary process (DSP). We first show that the results based on the approach should be interpreted with caution, as they are relatively more subject to the ‘pile-up problem’ than those based on the Bayesian approach. We then directly estimate and compare the two competing TSP and DSP models of real GDP within the Bayesian framework. Our empirical results suggest that a DSP model is preferred to a TSP model both in terms of in-sample fits and out-of-sample forecasts.

There has been a considerable amount of literature on binomial regression models that utilize well-known link functions, such as logistic, probit, and complementary log-log functions. The conventional binomial model is focused only on a single parameter representing one probability of success. However, we often encounter data for which two different success probabilities are of interest simultaneously. For instance, there are several offensive measures in baseball to predict the future performance of batters. Under these circumstances, it would be meaningful to consider more than one success probability. In this article, we employ a bivariate binomial distribution that possesses two success probabilities to conduct a regression analysis with random effects being incorporated under a Bayesian framework. Major League Baseball data are analyzed to demonstrate our methodologies. Extensive simulation studies are conducted to investigate model performances.