A power Series Distribution Involving the Double Factorial Emilio Gómez-déniz. | [Pakistan Journal of Statistics • 2016]

Author(s):

1. Emilio Gmez Dniz: Department of Quantitative Methods, University of Las Palmas de G.C. 35100 Las Palmas de Gran Canaria, SpainEstimation, Hazard, Natural Exponential Family, Over-dispersion, Power Series, Spain

Abstract:

This work proposes a one-parameter distribution with a formulation that involves the double factorial. The new distribution is intended to be competitive with the Poisson, geometric and negative binomial distributions and, like these, belongs to the natural exponential family of distributions (EFD) and to power series distributions. The new distribution is over-dispersed and appears to be unimodal or multimodal depending of the value of its parameter. The behaviour of the rate of successive probabilities is different from that of traditional discrete distributions, in which this rate is always increasing or decreasing. Furthermore, the hazard rate function also presents unusual behaviour in comparison with the classical distribution types mentioned above. We establish the normal approximation to the distribution proposed here. Estimation methods are considered, and applications show that the distribution works well when the data sets considered present two or more modal values.

Page(s): 457-473

DOI: DOI not available

Published: Journal: Pakistan Journal of Statistics, Volume: 32, Issue: 6, Year: 2016

Keywords:

Keywords are not available for this article.

References:

[1] Albano , M.,Amdeberhan , T.,Beyerstedt , E.,,2011,The error function,SCIENTIA. Series A: Mathematical Sciences,21,25-42

[2] Amidi , A.,,1976,A note on the moments of the generalized negative binomial distribution and on certain properties of this distribution,Siam J. Appl,34,223-224

[3] Arfken , G.,,1985,Mathematical Methods for Physicists,,,-

[4] Ayuso , M.,Bermúdez , L.,Santolino , M.,,2013,impeditivos utilizados en el cálculo de la indemnización básica por incapacidad temporal derivada de accidentes de tráfico,Riesgo 2013. Fundación Mapfre,,57-70

[5] Cardós , M.,Babiloni , E.,Estellés , S.,Guijarro , E.,,2013,A non parametric estimation of service level in a discrete context,Production Management and Engineering,2,47-52

[6] Bardwell , E.,,1960,,On Certain Characteristics of Some Discrete Distributions. Biometrika,47,473-475

[7] Bardwell , E.,Crow , E.L.,,1964,A two-parameter family of hyper-Poisson distributions,Journal of the American Statistical Association,59,133-141

[8] Bliss , C.I.,R.A.,,1953,Fitting the negative binomial distribution to biological data,Biometrics,9,176-200

[9] Claphama , R.,,1936,Over-dispersion in grassland communities and the use of statistical methods in plant ecology,Journal of Ecology,24,232-251

[10] Gould , H.,Quaintance , J.,,2012,Double fun with double factorials,Mathematics Magazine,85,177-192

[11] Johnson , N.L.,Kemp , A.W.,Kotz , S.,,2005,Univariate Discrete Distributions,Third Edition,,-

[12] Khatri , C.G.,,1959,On certain properties of power-series distributions,Biometrika,46,486-490

[13] KrishnaH.,PundirS.P.,,2009,Discrete Burr and Discrete Pareto Distributions,Statistical Methodology,6,177-188

[14] Lehmann , E.L.,Casella , G.,,1998,,Theory of Point Estimation,,-

[15] Min , A.,Czado , C.,,2010,Testing for zero-modification in count regression models,Statistica Sinica,20,323-341

[16] Meserve , B.E.,,1948,,American Mathematical Monthly,55,425-426

[17] Noack , A.,,1950,A class of random variables with discrete distributions,The Annals of Mathematical Statistics,21,127-132

[18] Papathanasiou , V.,,1993,Characterizations of power series and factorial series distributions,Sankhya,55,164-168

[19] Patil , G.P.,,1962,Maximum likelihood estimation for generalized power series distributions and its application to a truncated binomial distribution,Biometrika,49,227-237

[20] Ross , G.J.S.,Preece , D.A.,,1985,The negative binomial distribution,The Statistician,34,323-336

Citations

Downloads

Views