Similarities and Convergence in G-7 Cycles

Fiscal Divergence and Business Cycle Synchronization: Irresponsibility is Idiosyncratic

“…The literature gives only a few hints about the matter. Several authors argue that a world business cycle exists (e.g., Gerlach 1988;Lumsdaine and Prasad 2003;Darvas and Szapary 2007;Canova et al 2004), consistent with the absence of important asymmetries. Fatas and Mihov (2003a) studied discretionary fiscal policy for 91 countries and conclude (p. 1419) "governments that use fiscal policy aggressively induce significant macroeconomic instability" i.e., output volatility.…”

Revisiting the Impact of Oil Price Shocks on Macroeconomic Performance: An International Perspective

“…Abeysinghe and Forbes (2005) focus on the Asian financial crisis, studying the transmission mechanism through trade links. Canova et al (2007) study the dynamics and cross-country interdependencies in G-7 countries. Dees et al (2007) use a global VAR model to investigate the impact of external shocks on the euro area economy.…”

Bayesian Dynamic Tensor Regression

“…Another vector of regressors w t = vec W t ∈ R q may enter the regression (8) pre-multiplied (along mode-3) by a tensor D ∈ R m×n×q . Therefore, model (5) encompasses as a particular case also the panel VAR models of [16], [18], [17], provided that we make the same restriction on Σ.…”

“…This issue can be addressed by the indirect inclusion of parameter restrictions through a suitable specification of the corresponding prior distributions. In the unrestricted model (18) it would be necessary to define a prior distribution on the 4-order tensor B. The literature on tensor-valued distributions is limited to the elliptical family (e.g., [54]), which includes the tensor normal and tensor t. Both distributions do not easily allow for the specification of restrictions on a subset of the entries of the tensor, hampering the use of standard regularization prior distributions (such as shrinkage priors).…”

“…The specification of a tensor model with a tensor normal noise instead of a vector model (like a Gaussian VAR) has the advantage of being more parsimonious. By vectorising(18), we get the equivalent VAR…”