Volume 20, Issue 5

This study proposes methods for estimating Bayesian vector autoregressions (VARs) with a (semi-) automatic variable selection and an informative prior on the unconditional mean or steady-state of the system. We show that extant Gibbs sampling methods for Bayesian variable selection can be efficiently extended to incorporate prior beliefs on the steady-state of the economy. Empirical analysis, based on three major US macroeconomic time series, indicates that the out-of-sample forecasting accuracy of a VAR model is considerably improved when it combines both variable selection and steady-state prior information.

Latent factors estimated from panels of macroeconomic indicators are used to generate recession probabilities for the US economy. The focus is on current (rather than future) business conditions. Two macro factors are considered: (1) a dynamic factor estimated by maximum likelihood from a set of 4 monthly series; (2) the first of eight static factors estimated by principal components using a panel of 102 monthly series. Recession probabilities generated using standard probit, autoregressive probit, and Markov-switching models exhibit very different properties. Overall, a simple Markov-switching model based on the big data macro factor generates the sequence of out-of-sample class predictions that better approximates NBER recession months. Nevertheless, it is shown that the selection of the best performing model depends on the forecaster’s relative tolerance for false positives and false negatives.

Empirical macroeconomic research on business cycle typically filters economic time series in order to obtain cyclical components. This paper examines the effects of filtering data on the test for a linear autoregression against a threshold autoregression. Monte Carlo simulation shows that (1) filtering data in general reduces the power of the test, (2) the power is sensitive to the choice of filters and the specification of the trend and cyclical components, (3) regime-varying variance of the error term can affect the rejection frequency. Empirical evidences for cyclical asymmetry are provided for the quarterly US real GNP.

This paper investigates the inter-relationships between the gold price on the one hand, other precious metals (e.g. silver, palladium, platinum) and asset markets (e.g. stocks, bonds, crude oil) on the other hand. The econometric methodology relies on the Markov-switching BEKK model by Haas and Mittnik (2008) that captures time-varying correlations and bull-bear regimes for bivariate specifications. The model is applied to daily data from 1988 to 2013. The main results indicate that gold’s influence, through return and/or volatility spillovers, seems almost intact whatever the economic regime. Robustness checks of the statement that gold occupies a special place among commodities are provided under the form of a multi-asset portfolio management exercise.

Some economic models like the cash-in-advance model of money, the overlapping generations model and a model of credit with limited commitment may have the property that the dynamical system characterizing equilibria in the model are multi-valued going forward in time, but single-valued going backward in time. Such models or dynamical systems are said to have backward dynamics. In such instances, what does it mean for a dynamical system (set-valued) to be chaotic? Furthermore, under what conditions are such dynamical systems chaotic? In this paper, I provide a definition of chaos that is in the spirit of Li and Yorke for a dynamical system with backward dynamics. I utilize the theory of inverse limits to provide sufficient conditions for such a dynamical system to be Li-Yorke chaotic.

As scarce environmental resources necessarily put a constraint on population growth, we use more realistic population growth dynamics which contemplates a feedback mechanism between population growth rate and resource availability. We examine the local stability properties in overlapping generations resource economies which takes this feedback mechanism into account. The results indicate that Hopf bifurcation may arise without requiring logistic regeneration or unconventional constraints on parameter values. In particular, Hopf bifurcation is encountered under convex-concave dependence of carrying capacity on the resource availability.