A Class of Copula-Based Bivariate Poisson Time Series Models with Applications

A class of bivariate integer-valued time series models was constructed via copula theory. Each series follows a Markov chain with the serial dependence captured using copula-based transition probabilities from the Poisson and the zero-inflated Poisson (ZIP) margins. The copula theory was also used again to capture the dependence between the two series using either the bivariate Gaussian or “t-copula” functions. Such a method provides a flexible dependence structure that allows for positive and negative correlation, as well. In addition, the use of a copula permits applying different margins with a complicated structure such as the ZIP distribution. Likelihood-based inference was used to estimate the models’ parameters with the bivariate integrals of the Gaussian or t-copula functions being evaluated using standard randomized Monte Carlo methods. To evaluate the proposed class of models, a comprehensive simulated study was conducted. Then, two sets of real-life examples were analyzed assuming the Poisson and the ZIP marginals, respectively. The results showed the superiority of the proposed class of models.

[1]  Shan Hu,et al.  Dynamic Models for Time Series of Counts with a Marketing Application , 2015 .

[2]  A. Genz,et al.  Computation of Multivariate Normal and t Probabilities , 2009 .

[3]  Fukang Zhu,et al.  Flexible bivariate Poisson integer-valued GARCH model , 2019, Annals of the Institute of Statistical Mathematics.

[4]  Fukang Zhu,et al.  A new bivariate integer-valued GARCH model allowing for negative cross-correlation , 2018 .

[5]  LambertDiane Zero-inflated Poisson regression, with an application to defects in manufacturing , 1992 .

[6]  N. Diawara,et al.  Copula-based Markov zero-inflated count time series models with application , 2020, Journal of applied statistics.

[7]  Joe Harry Markov Models for Count Time Series , 2015 .

[8]  A. M. M. Shahiduzzaman Quoreshi,et al.  Bivariate Time Series Modeling of Financial Count Data , 2006 .

[9]  Andréas Heinen,et al.  Multivariate autoregressive modeling of time series count data using copulas , 2007 .

[10]  Nalini Ravishanker,et al.  Count Time Series: A Methodological Review , 2021 .

[11]  D. S. Young,et al.  Zero‐inflated modeling part II: Zero‐inflated models for complex data structures , 2020, WIREs Computational Statistics.

[12]  Kui Wang,et al.  A bivariate zero-inflated Poisson regression model to analyze occupational injuries. , 2003, Accident; analysis and prevention.

[13]  D. Karlis,et al.  Flexible Bivariate INAR(1) Processes Using Copulas , 2013 .

[14]  Aristidis K. Nikoloulopoulos,et al.  Coupling Couples with Copulas: Analysis of Assortative Matching on Risk Attitude , 2016, Economic Inquiry.

[15]  Konstantinos Fokianos,et al.  Multivariate Count Time Series Modelling , 2021, Econometrics and Statistics.

[16]  T. Baghfalaki,et al.  Pair copula construction for longitudinal data with zero-inflated power series marginal distributions , 2020, Journal of biopharmaceutical statistics.

[17]  P. Embrechts,et al.  Dependence modeling with copulas , 2007 .

[18]  Ostap Okhrin,et al.  Copulae: An overview and recent developments , 2021, WIREs Computational Statistics.

[19]  Li-Hsien Sun,et al.  Copula-Based Markov Models for Time Series: Parametric Inference and Process Control , 2020 .

[20]  M. Haugh,et al.  An Introduction to Copulas , 2016 .