Multivariate Poisson regression with covariance structure

Abstract

In recent years the applications of multivariate Poisson models have increased, mainly because of the gradual increase in computer performance. The multivariate Poisson model used in practice is based on a common covariance term for all the pairs of variables. This is rather restrictive and does not allow for modelling the covariance structure of the data in a flexible way. In this paper we propose inference for a multivariate Poisson model with larger structure, i.e. different covariance for each pair of variables. Maximum likelihood estimation, as well as Bayesian estimation methods are proposed. Both are based on a data augmentation scheme that reflects the multivariate reduction derivation of the joint probability function. In order to enlarge the applicability of the model we allow for covariates in the specification of both the mean and the covariance parameters. Extension to models with complete structure with many multi-way covariance terms is discussed. The method is demonstrated by analyzing a real life data set.

Spatio-Temporal Disease Surveillance

“…We model the trends in the counts using transitional Poisson regression model and assume that if the counts are incontrol then the one day ahead forecast errors are uncorrelated. In situations where spatial correlation persists in the forecast errors for in-control situations, an alternative approach is models which account for this spatial correlation, such as seeming unrelated Poisson regression models (King, 1989, andGrijalva, T., Bohara, et al 2003), and multivariate Poisson regression model Meligkotsidou, 2005, andBermudez andKarlis, 2011). However, here we model each of the 1600 cells separately using Poisson regression models very similar to that used in Sparks et al (2010a) and Sparks et al (2011a) that use explanatory variables: logarithm of lag counts plus 1, day-of-the-week, public holidays and harmonics.…”

A Multivariate Poisson Model of Consumer Choice in a Multi-Airport Region

“…The unknown parameters are estimated by an expectation maximization (EM) algorithm [4]. The EM algorithm is used for finding maximum likelihood estimates of probabilistic model parameters where the underlying data are unobservable.…”

“…4 Goodness of Fit is the scaled deviance. It is a chi-square divided by the degrees of freedom and with an expected value of one.…”

“…The fundamental difference between earlier work and that presented here is the difference between allocation modeling and modeling at the extensive margin. Until recently the use of multivariate Poisson regression was not an option [4]. An expectation maximization algorithm is used to estimate the parameters of a multivariate Poisson model of consumer decisions.…”

Modelling Dependence in a Ratemaking Procedure with Multivariate Poisson Regression Models

“…Boucher et al, 2007 andBermúdez, 2009). The first one, which we call the "common covariance model", has been defined in Tsionas (2001) and the second one, the "full covariance model", in Karlis and Meligkotsidou (2005). In addition, here we extend these models with their zero-inflated variants.…”