A0226
Title: Quantile spillover indexes: Simulation-based evidence, confidence intervals and a decomposition
Authors: Massimiliano Caporin - University of Padova (Italy) [presenting]
Giovanni Bonaccolto - University of Enna Kore (Italy)
Jawad Shahzad - Montpellier Business School (France)
Abstract: Quantile-spillover indexes have recently become popular for analyzing tail interdependence. An extensive simulation study shows that the estimation of spillover indexes is affected by a positive distortion when the parameters of the underlying fitted models are not evaluated with respect to their statistical significance. The distortion is reduced by filtering out non-significant parameters and also for increasing sample sizes, thanks to the consistency of estimators, but is not fully disappearing due to type I error. Another step is introducing a simulation-based approach to recovering confidence intervals from quantile spillover indexes. In addition, an algebraic decomposition of quantile spillover is put forward, separating the dynamic interdependence from the contemporaneous interdependence (due to residual correlation). Empirical evidence shows that distortions in real data are sizable, and the decomposition points out that most of the spillover is due to contemporary effects. All of the results extend and are confirmed for the spillover index of a prior study.
