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.

Key words： 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