Computing with bivariate COM-Poisson model under different copulas

Abstract Bivariate counts are collected in many sectors of research but the analysis of such data is often challenging because each series of counts may exhibit different levels and types of dispersion. This paper addresses this problem by proposing a flexible bivariate COM-Poisson model that may handle any combination of over-, equi- and under-dispersion at any levels. In this paper, the bivariate COM-Poisson is developed via Archimedean copulas. The Generalized Quasi-Likelihood (GQL) approach is used to estimate the unknown mean parameters in the copula-based bivariate COM-Poisson model while the dependence parameter is estimated using the copula likelihood. We further introduce a Monte Carlo experiment to generate bivariate COM-Poisson data under different dispersion levels. The performance of the GQL approach is assessed on the simulated data. The model is applied to analyze real-life epileptic seizures data.

[1]  Srinivas Reddy Geedipally,et al.  Application of the Conway-Maxwell-Poisson generalized linear model for analyzing motor vehicle crashes. , 2008, Accident; analysis and prevention.

[2]  Narayanaswamy Balakrishnan,et al.  Bivariate Conway-Maxwell-Poisson distribution: Formulation, properties, and inference , 2016, J. Multivar. Anal..

[3]  H. Joe.  Asymptotic efficiency of the two-stage estimation method for copula-based models , 2005 .

[4]  T. Minka,et al.  A useful distribution for fitting discrete data: revival of the Conway–Maxwell–Poisson distribution , 2005 .

[5]  D. Karlis,et al.  BIVARIATE INAR ( 1 ) MODELS , 2009 .

[6]  Henry Teicher,et al.  On the multivariate poisson distribution , 1954 .

[7]  Kirthi Kalyanam,et al.  Deconstructing Each Item's Category Contribution , 2007 .

[9]  Yuvraj Sunecher,et al.  Estimating the parameters of a BINMA Poisson model for a non-stationary bivariate time series , 2017, Commun. Stat. Simul. Comput..

[10]  Vandna Jowaheer,et al.  Comparing Joint GQL Estimation and GMM Adaptive Estimation in COM-Poisson Longitudinal Regression Model , 2013, Commun. Stat. Simul. Comput..

[11]  M. Sklar.  Fonctions de repartition a n dimensions et leurs marges , 1959 .

[12]  Fw Fred Steutel,et al.  Discrete analogues of self-decomposability and stability , 1979 .

[13]  Andréas Heinen,et al.  Multivariate modelling of time series count data: an autoregressive conditional poisson model , 2003 .

[14]  Gabriel Altmann,et al.  Towards a Theory of Word Length Distribution , 1994, J. Quant. Linguistics.

[15]  Maria Eduardo da Silva,et al.  Difference Equations for the Higher‐Order Moments and Cumulants of the INAR(1) Model , 2004 .

[16]  Sharad Borle,et al.  Computing with the COM-Poisson distribution , 2003 .

[17]  A. M'Kendrick.  Applications of Mathematics to Medical Problems , 1925, Proceedings of the Edinburgh Mathematical Society.

[18]  P. Thall,et al.  Some covariance models for longitudinal count data with overdispersion. , 1990, Biometrics.

[19]  B. Sutradhar.  An Overview on Regression Models for Discrete Longitudinal Responses , 2003 .

[20]  Y. Sunecher,et al.  Modelling a non-stationary BINAR(1) Poisson process , 2016 .

[21]  C. Genest,et al.  A Primer on Copulas for Count Data , 2007, ASTIN Bulletin.

[22]  Pravin K. Trivedi,et al.  Copula Modeling: An Introduction for Practitioners , 2007 .

[23]  Dimitris Karlis,et al.  Regression in a copula model for bivariate count data , 2010 .

[24]  E. McKenzie,et al.  Some ARMA models for dependent sequences of poisson counts , 1988, Advances in Applied Probability.

[25]  Srinivas Reddy Geedipally,et al.  Extension of the Application of Conway‐Maxwell‐Poisson Models: Analyzing Traffic Crash Data Exhibiting Underdispersion , 2010, Risk analysis : an official publication of the Society for Risk Analysis.

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

[27]  N. L. Johnson,et al.  Discrete Multivariate Distributions , 1998 .

[28]  F. Famoye,et al.  Bivariate generalized Poisson distribution with some applications , 1995 .

[29]  Klaus L. P. Vasconcellos,et al.  Improved estimation for Poisson INAR(1) models , 2015 .

[30]  Siddharth S. Singh,et al.  A Generalized Framework for Estimating Customer Lifetime Value When Customer Lifetimes are Not Observed , 2007 .

[31]  Emad-Eldin A. A. Aly,et al.  Stationary solutions for integer-valued autoregressive processes , 2005, Int. J. Math. Math. Sci..

[32]  Christian H. Weiß,et al.  Thinning operations for modeling time series of counts—a survey , 2008 .

[33]  Dimitris Karlis,et al.  Modeling Multivariate Count Data Using Copulas , 2009, Commun. Stat. Simul. Comput..

[34]  Siem Jan Koopman,et al.  Intraday Stock Price Dependence Using Dynamic Discrete Copula Distributions , 2015 .

[35]  R. Nelsen.  An Introduction to Copulas , 1998 .

[36]  Remarks on asymptotic efficient estimation for regression effects in stationary and nonstationary models for panel count data , 2014 .

[37]  Sharad Borle,et al.  A Model of the Joint Distribution of Purchase Quantity and Timing , 2003 .

[38]  Karl-Heinz Best.  Probability Distributions of Language Entities* , 2001, J. Quant. Linguistics.

[39]  Dimitris Karlis,et al.  A bivariate INAR(1) process with application , 2011 .

[40]  Joseph B. Kadane,et al.  The Timing of Bid Placement and Extent of Multiple Bidding: An Empirical Investigation Using eBay Online Auctions , 2006, math/0609194.

[41]  Dimitris Karlis,et al.  Some properties of multivariate INAR(1) processes , 2013, Comput. Stat. Data Anal..

[42]  J. S. Maritz,et al.  Note on a certain family of discrete distributions , 1952 .

[43]  Aristidis K. Nikoloulopoulos.  Copula-Based Models for Multivariate Discrete Response Data , 2013 .

[44]  Yuvraj Sunecher,et al.  Analyzing Multivariate Cross-Sectional Poisson Count Using a Quasi-Likelihood Approach: The Case of Trivariate Poisson , 2016 .