Title: Time-varying Poisson autoregression
Authors: Giovanni Angelini - University of Bologna (Italy) [presenting]
Abstract: A new time-varying econometric model is proposed, called Time-Varying Poisson AutoRegressive (TV-PAR), with is suited to forecast time series of counts. We show that the score-driven framework is particularly suitable to recover the evolution of time-varying parameters and provides the required flexibility to model and forecast time series of counts characterized by convoluted nonlinear dynamics and structural breaks. We study the asymptotic properties of the TV-PAR model and prove that, provided some mild conditions, maximum likelihood estimation (MLE) yields strongly consistent and asymptotically normal parameter estimates. Finite-sample performance and forecasting accuracy are evaluated through Monte Carlo simulations. The empirical usefulness of the time-varying specification of the proposed TV-PAR model is shown by analyzing the number of monthly corporate defaults.