The zero inflated negative binomial-beta exponential (ZINB-BE) distribution is developed, it is an alternative distribution for the excessive zero counts with overdispersion. The ZINB-BE distribution is a mixture of two distributions which are Bernoulli and negative binomial-beta exponential distributions. In this work, some characteristics of the proposed distribution are presented, such as, mean and variance. The maximum likelihood estimation is applied to parameter estimation of the proposed distribution. Finally some results of Monte Carlo simulation study, it seems to have high-efficiency when the sample size is large.
REFERENCES
1.
D. R.
Winkelmann
, Econometric Analysis of Count Data
(Springer
Berlin Heidelberg
, 2000
).2.
N.
Johnson
and S.
Kotz
, Discrete distributions, Distributions in statistics
(Houghton Mifflin
, 1969
).3.
D.
Lambert
, Technometrics
34
, 1
–14
(1992
).4.
M. J.
Campbell
, D.
Machin
, and C.
D’Arcangues
, Statistics in medicine
10
, 241
–254
(1991
).5.
D.
Böhning
, E.
Dietz
, P.
Schlattmann
, L.
Mendonca
, and U.
Kirchner
, Journal of the Royal Statistical Society: Series A (Statistics in Society)
162
, 195
–209
(1999
).6.
D. A.
Freund
, T. J.
Kniesner
, and A. T.
LoSasso
, Economics Letters
62
, 7
–12
(1999
).7.
V.
Shankar
, J.
Milton
, and F.
Mannering
, Accident Analysis & Prevention
29
, 829
–837
(1997
).8.
S.-P.
Miaou
, Accident Analysis & Prevention
26
, 471
–482
(1994
).9.
C.
Pudprommarat
, W.
Bodhisuwan
, and P.
Zeephongsekul
, Journal of Applied Sciences
12
, 1853
–1858
(2012
).10.
R Core Team
, R: A Language and Environment for Statistical Computing
, R Foundation for Statistical Computing
, Vienna, Austria
(2016
).
This content is only available via PDF.
© 2017 Author(s).
2017
Author(s)
You do not currently have access to this content.
