Multivariate score-driven models for count time series with application to credit risk

This paper develops a new multivariate model for count time series, in which the time-varying intensity parameter determining the probability that an event occurs evolves according to a general autoregressive score (GAS) specification. The model is particularly suitable to study shock propagation channels between different economic sectors or markets. Indeed, the interdependence between event counts arises from the effect of shocks in the number of events that occurred in a sector on the probability that new events occur in other sectors. By applying the model to daily CDS spread data relative to a sample of European companies, we find significant within and cross-sector effects. In particular, the Financial and Energy sectors are those whose credit risk events impact others the most, while the sectors most affected by events in other markets turn out to be ICT and Trade.