G.H. Dash, Jr. and N. Kajiji (USA)
Financial visualization, volatility spillovers, neural networks.
In this paper we extend prior efforts to engineer an efficient mapping of volatility transmission across various western- and central-European government bond markets. The closed-form derivation of the Bayesian-enhanced regularization parameter embodied by the Kajiji-4 RBF ANN is known to produce an efficient minimization of the ill-effects of multi-collinearity while attaining maximum smoothness in nonparametric time series analysis. This computational innovation provides the raison d être for a comparative re-examination of the bond volatility spillover effects obtained from contemporary parametric-based conditional volatility investigations. The research proposes a two step methodology that first engineers an efficient ANN modelling of the latent EGARCH effects in the target financial time-series which is then followed by a focus on comparative modelling efficiency using both linear and nonlinear residual return diagnostics. We report substantial evidence to support the ex-ante expectation for the Kajiji-4 RBF ANN to produce residuals that are devoid of latent economic covariance and conditional volatility effects. Policy inferences generated from the model s weights clearly corroborate reported findings of a weak US spillover effect into established European bond markets while offering new insights into the inverted volatility spillover effects attributed to US support for select developing central-European economies.
Important Links:
Go Back
