COM-negative binomial distribution: modeling overdispersion and ultrahigh zero-inated count data

Abstract

We focus on the COM-type negative binomial distribution with three parameters, which belongs to COM-type (a, b, 0) class distributions and family of equilibrium distributions of arbitrary birth-death process. Besides, we show abundant distributional properties such as overdispersion and underdispersion, log-concavity, log-convexity (infinite divisibility), pseudo compound Poisson, stochastic ordering, and asymptotic approximation. Some characterizations including sum of equicorrelated geometrically distributed random variables, conditional distribution, limit distribution of COM-negative hypergeometric distribution, and Stein's identity are given for theoretical properties. COM-negative binomial distribution was applied to overdispersion and ultrahigh zero-inated data sets. With the aid of ratio regression, we employ maximum likeli-hood method to estimate the parameters and the goodness-of-fit are evaluated by the discrete Kolmogorov-Smirnov test.

Keywords

Overdispersion / zero-inated data / infinite divisibility / Stein's characterization / discrete Kolmogorov-Smirnov test

References

[1] Arnold T B, Emerson J W. Nonparametric goodness-of-fit tests for discrete null distributions. The R Journal, 2011, 3(2): 34–39

[2] Böhning D. Ratio plot and ratio regression with applications to social and medical sciences. Statist Sci, 2016, 31(2): 205–218

[3] Borges P, Rodrigues J, Balakrishnan N, Bazán J. A COM-Poisson type generalization of the binomial distribution and its properties and applications. Stat Probab Lett, 2014, 87: 158–166

[4] Brown T C, Phillips M J. Negative binomial approximation with Stein's method. Methodol Comput Appl Probab, 1999, 1(4): 407–421

[5] Brown T C, Xia A. Stein's method and birth-death processes. Ann Probab, 2001, 29(3): 1373–1403

[6] Chakraborty S, Ong S H. A COM-Poisson-type generalization of the negative binomial distribution. Comm Statist Theory Methods, 2016, 45(14): 4117–4135

[7] Chakraborty S, Imoto T. Extended Conway-Maxwell-Poisson distribution and its properties and applications. J Stat Distrib App, 2016, 3: 5

[8] Conway R W, Maxwell W L. A queuing model with state dependent service rates. J Industrial Engineering, 1962, 12(2): 132–136

[9] Denuit M, Maréchal X, Pitrebois S, Walhin J-F. Actuarial Modelling of Claim Counts: Risk Classification, Credibility and Bonus-Malus Systems.Chichester: John Wiley & Sons, 2007

[10] Gómez-Déniz E, Sarabia José María E, Calderín-Ojeda. A new discrete distribution with actuarial applications. Insurance Math Econom, 2011, 48(3): 406–412

[11] Gómez-Déniz E, Calderín-Ojeda E. Unconditional distributions obtained from conditional specification models with applications in risk theory. Scand Actuar J, 2014, (7): 602–619

[12] Gupta R C, Sim S Z, Ong S H. Analysis of discrete data by Conway-Maxwell Poisson distribution. Adv Stat Anal, 2014, 98(4): 327–343

[13] Haberman S J. A warning on the use of chi-squared statistics with frequency tables with small expected cell counts. J Amer Statist Assoc, 1988, 83(402): 555–560

[14] Ibragimov I A. On the composition of unimodal distributions. Theory Probab Appl, 1956, 1(2): 255–260

[15] Imoto T. A generalized Conway-Maxwell-Poisson distribution which includes the negative binomial distribution. Appl Math Comput, 2014, 247: 824–834

[16] Johnson N L, Kemp A W, Kotz S. Univariate Discrete Distributions.3rd ed.Hoboken: John Wiley & Sons, 2005

[17] Kadane J B. Sums of possibly associated Bernoulli variables: the Conway-Maxwell-Binomial distribution. Bayesian Anal, 2016, 11(2): 403–420

[18] Kagan A M, Rao C R, Linnik Y V. Characterization Problems in Mathematical Statistics.New York: Wiley

[19] Kaas R, Goovaerts M, Dhaene J, Denuit M. Modern Actuarial Risk Theory Using R.2nd ed. Berlin; Springer, 2008

[20] Kokonendji C C, Mizere D, Balakrishnan N. Connections of the Poisson weight function to overdispersion and underdispersion. J Statist Plann Inference, 2008, 138(5): 1287–1296

[21] Liu Q, Lee J, Jordan M. A kernelized Stein discrepancy for goodness-of-fit tests. International Conference on Machine Learning, 2016, 276–284

[22] Meintanis S G, Nikitin Y Y. A class of count models and a new consistent test for the Poisson distribution. J Statist Plann Inference, 2008, 138(12): 3722–3732

[23] Patil G P, Seshadri V. Characterization theorems for some univariate probability distributions. J R Stat Soc Ser B Stat Methodol, 1964, 26: 286–292

[24] Rao C R, Rubin H. On a characterization of the Poisson distribution. Sankhyā, 1964, 32(2-3): 295–298

[25] Rényi A. On measures of entropy and information. In: Proceedings of the Fourth Berkeley Symposium on Mathematical Statistics and Probability, Vol 1. Contributions to the Theory of Statistics. The Regents of the University of California, 1961, 547–561

[26] Ramalingam S, Jagbir S. A characterization of the logarithmic series distribution and its application. Comm Statist Theory Methods, 1984, 13(7): 865–875

[27] Rodrigues J, de Castro M, Cancho V G, Balakrishnan N. COM-Poisson cure rate survival models and an application to a cutaneous melanoma data. J Statist Plann Inference, 2009, 139(10): 3605–3611

[28] Sellers K F, Borle S, Shmueli G. The COM-Poisson model for count data: a survey of methods and applications. Appl Stoch Models Bus Ind, 2012, 28(2): 104–116

[29] Shaked M, Shanthikumar J G. Stochastic Orders.Berlin: Springer, 2007

[30] Shanbhag D N. An extension of the Rao-Rubin characterization of the Poisson distribution. J Appl Probab, 1977, 14(3): 640–646

[31] Shmueli G, Minka T P, Kadane J B, Borle S, Boatwright P. A useful distribution for fitting discrete data: revival of the Conway-Maxwell-Poisson distribution. J R Stat Soc Ser C Appl Stat, 2005, 54(1): 127–142

[32] Steutel F W. Preservation of Infinite Divisibility under Mixing and Related Topics. Math Centre Tracts, 33.Amsterdam: Mathematisch Centrum, 1970

[33] Steutel F W, van Harn K. Infinite Divisibility of Probability Distributions on the Real Line.Boca Raton: CRC Press, 2003

[34] Tsallis C. Possible generalization of Boltzmann-Gibbs statistics. J Stat Phys, 1988, 52(1-2): 479–487

[35] Temme N M. Special Functions: An Introduction to the Classical Functions of Mathematical Physics.New York: John Wiley & Sons, 2011

[36] Wang L M, Lei Y L. Simulation and EM algorithm for the distribution of number of claim in the heterogeneous portfolio. Commun Appl Math Comput Sci, 2000, 14(2): 71–78

[37] Willmot G E. The Poisson-inverse Gaussian distribution as an alternative to the negative binomial. Scand Actuar J, 1987, (3-4): 113–127

[38] Wimmer G, Köhler R, Grotjahn R, Altmann G. Towards a theory of word length distribution. J Quant Linguist, 1994, 1(1): 98–106

[39] Wimmer G, Altmann G. Thesaurus of Univariate Discrete Probability Distributions.Essen: Stamm, 1999

[40] Zhang H, Liu Y, Li B. Notes on discrete compound Poisson model with applications to risk theory. Insurance Math Econom, 2014, 59: 325–336

[41] Zhang H, Li B. Characterizations of discrete compound Poisson distributions. Comm Statist Theory Methods, 2016, 45(22): 6789–6802

[42] Zhang H, Li B, Kerns G J. A characterization of signed discrete infinitely divisible distributions. Studia Sci Math Hung, 2017, 54(4): 446–470

[43] Zhang H, Jia J. Elastic-net regularized high-dimensional negative binomial regression: consistency and weak signals detection. arXiv: 1712.03412

[44] Zygmund A. Trigonometric Series.Cambridge: Cambridge University Press, 2002

RIGHTS & PERMISSIONS

2018 Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature