Title: A Bayesian graphical VAR model for yield curve fluctuations
Authors: Monica Billio - University of Venice (Italy) [presenting]
Andrea Berardi - University of Venice (Italy)
Roberto Casarin - University Ca' Foscari of Venice (Italy)
Abstract: Yield curve fluctuations across different currency areas are generally highly interrelated. However, both the contemporaneous causal relationships and the temporal dependence structure vary over time. We document the time-varying behaviour of the degree of connectedness among yield changes in seven currency areas (Australia, Canada, Germany, Japan, Switzerland, UK and US). We decompose yields into expected short rates and term premia using a Gaussian ATSM integrated with long-term yield expectations and analyse the contribution of those components to global yield co-movements and connectedness. We find that the dependence structure of both yields and their components can be significantly different for short and long maturities. The empirical analysis is based on a Bayesian graphical VAR model, where the contemporaneous and temporal causal structures of the structural VAR are represented by two different graphs and an efficient Markov chain Monte Carlo algorithm is used to estimate jointly the two causal structures and the parameters of the reduced-form VAR model. When representing bond yields as a network, the increased system fragility is reflected by a degree distribution which is symmetric and has thinner tails, whereas asymmetry and fat tails suggest that there is heterogeneity in the linkages among countries.