A1631
Title: Score-driven generalized Poisson model
Authors: Giulia Carallo - Ca' Foscari University of Venice (Italy) [presenting]
Roberto Casarin - University Ca' Foscari of Venice (Italy)
Dario Palumbo - Ca' Foscari University of Venice (Italy)
Abstract: A new score driven model for integer data is introduced. In particular, we introduce a dynamic conditional score model where the series has Generalized Poisson conditional distribution (GP-DCS), for the location and scale parameters. We provide a Bayesian inference framework and an efficient posterior approximation procedure based on Markov Chain Monte Carlo. An application to fire data shows that the proposed DCS model is well suited for capturing persistence in the conditional moments and in the over-dispersion feature of the data.
