New exponential dispersion models for count data: the ABM and LM classes

In their fundamental paper on cubic variance functions (VFs), Letac and Mora (The Annals of Statistics, 1990) presented a systematic, rigorous and comprehensive study of natural exponential families (NEFs) on the real line, their characterization through their VFs and mean value parameterization. They presented a section that for some reason has been left unnoticed. This section deals with the construction of VFs associated with NEFs of counting distributions on the set of nonnegative integers and allows to find the corresponding generating measures. As EDMs are based on NEFs, we introduce in this paper two new classes of EDMs based on their results. For these classes, which are associated with simple VFs, we derive their mean value parameterization and their associated generating measures. We also prove that they have some desirable properties. Both classes are shown to be overdispersed and zero inflated in ascending order, making them as competitive statistical models for those in use in both, statistical and actuarial modeling. To our best knowledge, the classes of counting distributions we present in this paper, have not been introduced or discussed before in the literature. To show that our classes can serve as competitive statistical models for those in use (e.g., Poisson, Negative binomial), we include a numerical example of real data. In this example, we compare the performance of our classes with relevant competitive models.

[1]  José María Sarabia,et al.  A new discrete distribution with actuarial applications , 2011 .

[2]  Gérard Letac,et al.  Natural Real Exponential Families with Cubic Variance Functions , 1990 .

[3]  Emilio Gómez-Déniz,et al.  The discrete Lindley distribution: properties and applications , 2011 .

[4]  C. Morris.  Natural Exponential Families with Quadratic Variance Functions , 1982 .

[5]  C. C. Kokonendji,et al.  On Strict Arcsine Distribution , 2004 .

[6]  Ad Ridder,et al.  Exponential dispersion models for overdispersed zero-inflated count data , 2020, Commun. Stat. Simul. Comput..

[7]  Modelling claim number using a new mixture model: negative binomial gamma distribution , 2016 .

[8]  B. Jørgensen.  Exponential Dispersion Models , 1987 .

[9]  C. C. Kokonendji,et al.  A Property of Count Distributions in the Hinde–Demétrio Family , 2008 .

[10]  Ronald Lee,et al.  Modeling and forecasting U. S. mortality , 1992 .

[11]  Michael J. LuValle,et al.  Generalized Poisson Distributions: Properties and Applications , 1990 .

[12]  Bent Jørgensen.  The Theory of Exponential Dispersion Models and Analysis of Deviance , 1992 .

[13]  Mourad Ismail,et al.  Approximation Operators, Exponential, q-Exponential, and Free Exponential Families , 2005 .

[14]  J. F. C. Kingman,et al.  Information and Exponential Families in Statistical Theory , 1980 .

[15]  Steven Haberman,et al.  A cohort-based extension to the Lee-Carter model for mortality reduction factors , 2006 .

[16]  Ad Ridder,et al.  Monte Carlo Methods for Insurance Risk Computation , 2018, International Journal of Statistics and Probability.

[17]  On the two-parameter Bell–Touchard discrete distribution , 2020, Communications in Statistics - Theory and Methods.

[18]  Michel Denuit,et al.  A New Distribution of Poisson-Type for the Number of Claims , 1997, ASTIN Bulletin.

[19]  C. C. Kokonendji,et al.  On Poisson-exponential-Tweedie models for ultra-overdispersed count data , 2020, AStA Advances in Statistical Analysis.

[20]  Leon M. Hall,et al.  Special Functions , 1998 .

[21]  Célestin C. Kokonendji,et al.  Discrete dispersion models and their Tweedie asymptotics , 2014, 1409.7482.

[22]  C. C. Kokonendji,et al.  On the mean value parametrization of natural exponential families — a revisited review , 2017 .

[23]  C. C. Kokonendji,et al.  On Hinde-Demétrio Regression Models for Overdispersed Count Data , 2007 .

[24]  R. Löhner.  Approximation of Operators , 2008 .

[25]  A Suitable Discrete Distribution for Modelling Automobile Claim Frequencies , 2016 .

[26]  H. Bakouch,et al.  A new infinitely divisible discrete distribution with applications to count data modeling , 2019 .

[27]  Matti Ruohonen.  On A Model for the Claim Number Process , 1988, ASTIN Bulletin.

[28]  C. C. Kokonendji,et al.  Extended Poisson–Tweedie: Properties and regression models for count data , 2016, 1608.06888.

[29]  A. W. Kemp,et al.  Generalized Poisson Distributions: Properties and Applications. , 1992 .

[30]  O. Barndorff-Nielsen.  Information and Exponential Families in Statistical Theory , 1980 .

[31]  John Hinde,et al.  Overdispersion: models and estimation , 1998 .

[32]  Shaul K. Bar-Lev,et al.  Reproducibility and natural exponential families with power variance functions , 1986 .

[33]  Gordon E. Willmot,et al.  The Poisson-Inverse Gaussian distribution as an alternative to the negative binomial , 1987 .

[34]  Harry Joe,et al.  Modelling species abundance using the Poisson–Tweedie family , 2011 .

[35]  Felix Famoye,et al.  Lagrangian Probability Distributions , 2005 .

[36]  Célestin C. Kokonendji,et al.  Some discrete exponential dispersion models: poisson-Tweedie and Hinde-Demetrio classes , 2004 .