Hierarchical Markov-Switching Models for Multivariate Integer-valued Time-series

29 Pages Posted: 16 Dec 2017 Last revised: 5 Dec 2019

See all articles by Leopoldo Catania

Leopoldo Catania

Aarhus University - School of Business and Social Sciences; Aarhus University - CREATES

Roberto Di Mari

University of Catania - Department of Economics and Quantitative Methods

Date Written: December 13, 2017

Abstract

We propose a new flexible dynamic model for multivariate nonnegative integer–valued time–series. Observations are assumed to depend on the realization of two unobserved integer–valued stochastic variables which control for the time– and cross–dependence of the data. We provide conditional and unconditional (cross)–moments implied by the model, as well as the limiting distribution of the series. An Expectation–Maximization algorithm for maximum likelihood estimation of the model’s parameters is derived. A Monte Carlo experiment investigates the finite sample properties of our estimation methodology. Constrained specifications of the model are recovered by modifying the assumptions about the dependence structure of the latent variables and model identification is discussed accordingly. An application by means of a crime data set from the New South Wales (NSW) Bureau Of Crime Statistics And Research with observations spanning beyond 20 years is reported. Results indicate that the proposed approach provides a good description of the conditional distribution of crime records, outperforming the standard hidden Markov model

Keywords: Hidden Markov Model, Mixture Model, Hierarchical Model, NSW Crime Data

Suggested Citation

Catania, Leopoldo and Di Mari, Roberto, Hierarchical Markov-Switching Models for Multivariate Integer-valued Time-series (December 13, 2017). Available at SSRN: https://ssrn.com/abstract=3087201 or http://dx.doi.org/10.2139/ssrn.3087201

Leopoldo Catania (Contact Author)

Aarhus University - School of Business and Social Sciences ( email )

Fuglesangs Allé 4
Aarhus V, DK-8210
Denmark
+4587165536 (Phone)

HOME PAGE: http://pure.au.dk/portal/en/leopoldo.catania@econ.au.dk

Aarhus University - CREATES ( email )

School of Economics and Management
Building 1322, Bartholins Alle 10
DK-8000 Aarhus C
Denmark

Roberto Di Mari

University of Catania - Department of Economics and Quantitative Methods ( email )

Corso Italia, 55
95129 Catania
Italy

HOME PAGE: http://www.datasciencegroup.unict.it/content/roberto-di-mari

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
119
Abstract Views
685
Rank
422,206
PlumX Metrics