Marshall-Olkin generalized Erlang-truncated exponential · PDF fileOkorie et al., Cogent...
Transcript of Marshall-Olkin generalized Erlang-truncated exponential · PDF fileOkorie et al., Cogent...
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
STATISTICS | RESEARCH ARTICLE
Marshall-Olkin generalized Erlang-truncated exponential distribution: Properties and applicationsIdika E. Okorie1*, Anthony C. Akpanta2 and Johnson Ohakwe3
Abstract: This article introduces the Marshall–Olkin generalized Erlang-truncated exponential (MOGETE) distribution as a generalization of the Erlang-truncated exponential (ETE) distribution. The hazard rate of the new distribution could be increasing, decreasing or constant. Explicit-closed form mathematical expressions of some of the statistical and reliability properties of the new distribution were given and the method of maximum likelihood estimation was used to estimate the model parameters. The usefulness and flexibility of the new distribution was illustrated with two real and uncensored lifetime data-sets. The MOGETE distribution with a smaller goodness of fit statistics always emerged as a better candidate for the data-sets than the ETE, Exp Fréchet and Exp Burr XII distributions.
Subjects: Science; Mathematics & Statistics; Physical Sciences
Keywords: Erlang-truncated exponential distribution; Marshall-Olkin; reliability; failure rate; AIC
1. IntroductionThe exponential distribution is about the simplest distribution in terms of expression and analytical tractability and widely used in reliability engineering. There is no doubt that the wide applicability of the exponential distribution even in inappropriate scenarios is motivated by its simplicity. However,
*Corresponding author: Idika E. Okorie, School of Mathematics, University of Manchester, Manchester M13 9PL, UK E-mail: [email protected]
Reviewing editor:Hiroshi Shiraishi, Keio University, Japan
Additional information is available at the end of the article
ABOUT THE AUTHORSIdika E. Okorie obtained his BSc in Statistics (2009) from the Abia State University, Nigeria and MSc in Statistics (Financial Statistics) (2014) from the University of Manchester, UK. He is presently studying for a PhD in Statistics at the University of Manchester, UK. His research interest includes Statistical modelling; and Distribution theory with applications.
Anthony C. Akpanta obtained his PhD in Statistics (2008) from the Abia State University, Nigeria. He is currently working as an associate professor of Statistics at the Department of Statistics, Abia State University, Nigeria. His area of research includes Time series analysis and forecasting; Sampling methods and distribution theory.
Johnson Ohakwe obtained his PhD in Statistics (2009), MSc Statistics (2005) and BSc Statistics (1998) from the Abia State University, Nigeria. He is currently a Senior lecturer at the Department of Mathematics, Computing and Physical sciences, Federal University Otuoke, Nigeria. His research interest include Time series and Forecasting and Distribution theory.
PUBLIC INTEREST STATEMENTThe simple structure of the proposed MOGETE distribution makes it easy to work with analytically and in practical situations, it provides a better fit to data-sets than some of the already existing distributions like the Erlang-truncated exponential distribution, Exponentiated Fréchet distribution and the Exponentiated Burr XII distribution. In addition to the example of possible applications of the new distribution as shown here, MOGETE distribution is suitable for modelling infant mortality rate and failure rate of some devices/equipments due to ageing.
Received: 08 November 2016Accepted: 15 January 2017First Published: 21 January 2017
Page 1 of 19
© 2017 The Author(s). This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.
Page 2 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
the exponential distribution has a major problem of constant failure/hazard rate property which makes it inappropriate for modelling data-sets from various complex life phenomena that may ex-hibit increasing, decreasing or bathtub hazard rate characteristics. El-Alosey (2007) extended the standard one parameter exponential distribution to a two parameter Erlang-truncated exponential (ETE) distribution. The pdf f (x) of the ETE distribution is given by
with cdf F(x)
and hazard rate function (hrf ) h(x)
where � is the shape parameter while � is the scale parameter. It is important to note that the ETE distribution has a constant hazard rate function.
The inability of the existing standard distributions to adequately model a variety of complex real data-sets; particularly, lifetime ones has stirred huge concern amongst distribution users and re-searchers alike and has summoned enormous research attention over the last two decades. Interestingly, tremendous research breakthroughs have been recorded by many researchers in their quest to the solution of the lack of fits limitation of the standard probability distributions. Among others is Marshall and Olkin (1997) who introduced the family of distributions that is known as the Marshall–Olkin extended/generalized distributions. The Marshall–Olkin’s technique of adding an ex-tra parameter to the original distribution has remarkably been known for its ability of producing more flexible and robust distributions that can represent a wide-ranging coverage of data-sets that emanates from a variety of complex phenomena. The Marshall–Olkin family of distributions can be obtained as follows,
It follows that F(x) = 1 − F(x) and
where G(x) and g(x) are the complementary cumulative density function (survival/reliability func-tion) and density function corresponding to the baseline distribution (original distribution).
A lot of standard probability distributions have been generalized by various researchers using the Marshall–Olkin procedure. For example, Ristić and Kundu (2015) introduced the Marshall–Olkin gen-eralized exponential distribution generalizing the exponentiated exponential distribution. Ghitany, Al-Hussaini, and Al-Jarallah (2005) introduced the Marshall–Olkin extended Weibull distribution as a generalization of the standard Weibull distribution. Ghitany (2005) introduced the Marshall–Olkin extended Pareto distribution as a generalization of the standard Pareto distribution. Ristić, Jose, and Ancy (2007) introduced the Marshall–Olkin extended gamma distribution as a generalization of the standard gamma distribution. Ghitany, Al-Awadhi, and Alkhalfan (2007) introduced the Marshall–Olkin extended Lomax distribution as a generalization of the standard Lomax distribution. Jose and Krishna (2011) introduced the Marshall–Olkin extended continuous uniform distribution as a gener-alization of the standard continuous uniform distribution. Al-Saiari, Baharith, and Mousa (2014) in-troduced the Marshall–Olkin extended Burr type XII distribution as a generalization of the standard Burr type XII distribution. Alizadeh et al. (2015) introduced the Marshall–Olkin extended
(1)f (x) = 𝛽(1 − e−𝜆)e−𝛽(1−e−𝜆)x; 0 ≤ x < ∞, 𝛽, 𝜆 > 0,
(2)F(x) = 1 − e−𝛽(1−e−𝜆)x; 0 ≤ x < ∞, 𝛽, 𝜆 > 0,
(3)h(x) = 𝛽(1 − e−𝜆); 𝛽, 𝜆 > 0,
(4)F(x) =𝛼G(x)
1 − (1 − 𝛼)G(x); −∞ < x < ∞; 0 < 𝛼 < ∞
(5)f (x) =𝛼g(x)
(1 − (1 − 𝛼)G(x)
)2 ; −∞ < x < ∞; 0 < 𝛼 < ∞
Page 3 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Kumaraswamy distribution as a generalization of the standard Kumaraswamy distribution. Gui (2013) introduced the Marshall–Olkin extended log-logistic distribution as a generalization of the standard log-logistic distribution. Pogány, Saboor, and Provost (2015) introduced the Marshall–Olkin extended exponential Weibull distribution generalizing the exponential Weibull distribution. Jose (2011) gave a comprehensive review of the Marshall–Olkin family of distributions and their applica-tions to reliability, time series and stress-strength analysis. For more extensive reviews of the Marshall–Olkin generalized family of distributions see, Nadarajah (2008) and Barreto-Souza, Lemonte, and Cordeiro (2013). Sandhya and Prasanth (2014) introduced the Marshall–Olkin extend-ed discrete uniform distribution as a generalizion of the standard discrete uniform distribution; etc.
Motivated by the idea of additional parameter for extra flexibility to the distribution, we introduce the three-parameter Marshall–Olkin generalized Erlang-truncated exponential (MOGETE) distribu-tion as a generalization of the standard two parameter ETE distribution. The importance of the new distribution is the ability of describing real data-sets with unimodal density as well as decreasing or increasing hazard rate function better than some already existing distributions as we show later. Hence, the MOGETE distribution has a superior fitting ability than the ETE distribution.
The remaining part of this article is organized as follows: Section 2 introduces the MOGETE distri-bution; Section 3 presents some reliability characteristics of the distribution such as the reliability function, hazard rate function and the mean residual life time; Section 4 presents some statistical properties of the new distribution such as the kth crude moment, mean, variance, coefficient of vari-ation, skewness, kurtosis, moment generating function, pth quantile function, Rényi entropy meas-ure of the new distribution and the distribution of order statistics of the distribution; Section 5 proposes the estimation of the distribution parameters through the method of maximum likelihood estimation; Section 6 presents the application of the new distribution to two real data sets; Section 7 presents the discussion of results and lastly and Section 8 is the conclusion of the study.
2. The MOGETE distributionThe cdf of the MOGETE distribution is given by
with the corresponding pdf as
where � and � are the shape parameters and � is the scale parameter.
Theorem 2.1 The MOGETE distribution with pdf in Equation (7) is identifiable i.e. ∃f (x;�1, �1, �1) = f (x; �2, �2, �2) as we show now.
Proof By setting
we have that
∀ x ≥ 0, the above equality is true ∀ �1 = �2, �1 = �2, and �1 = �2. ✷
(6)F(x) = 1 − F(x) =1 − e−𝛽(1−e
−𝜆)x
1 − (1 − 𝛼)e−𝛽(1−e−𝜆)x; 0 ≤ x < ∞, 𝛼 > 0, 𝛽, 𝜆 > 0
(7)f (x) = 𝛼𝛽(1 − e−𝜆)e−𝛽(1−e−𝜆)x
{1 − (1 − 𝛼)e−𝛽(1−e
−𝜆)x}−2
; 0 ≤ x < ∞, 𝛼 > 0, 𝛽, 𝜆 > 0
f (x; �1, �1, �1) = f (x; �2, �2, �2)
log(�1 − �2)3 + log(�1 − �2) + �1 − �2 − 3[�1(1 − e
−�1 ) − �2(1 − e−�2 )]x = 0
Page 4 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Theorem 2.2 If a random variable say X is distributed according to the MOGETE distribution then the shape of its pdf as x → ∞ is decreasing when � ≤ 2 and unimodal when 𝛼 > 2.
Proof The first derivative of the pdf f ′ in Equation (7) is given by
Setting f � = 0 gives the critical point x0 at which the pdf is maximized. x0 is the root of the equation which is given by
this implies that as x → ∞ and 𝛼 > 2 there exists some x < x0 such that f (x) > 0 and some x > x0 such that f (x) < 0, hence; f(x) has a single mode at x0. Now, it makes sense to conclude that the pdf have decreasing shape as the only alternative shape when � ≤ 2; since, f �(x0) ≠ 0 and both conditions (𝛼 ≤ 2 and 𝛼 > 2) cannot be jointly satisfied in each case (monotonic decreasing and unimodality). ✷
The asymptotic behaviour of the pdf of the MOGETE distribution is f (0) = �(1 − e−�)∕� and f (∞) = 0.
3. Reliability analysis with the MOGETE distributionIn this section, we present some reliability characteristics of the MOGETE distribution that is neces-sary for reliability analysis, they are: the reliability (survival) function F(x) (R(x)), the hazard rate function h(x) and the mean residual life time M(t).
3.1. Reliability functionThe reliability function R(x) is an important tool in reliability analysis for characterizing life phenom-enon. R(x) is mathematically expressed as R(x) = 1 − F(x). Under certain predefined conditions the reliability function generally gives the estimated probability that a system will operate without fail-ure until a specified time x. The reliability function of the MOGETE distribution is given by
For various parameter values R(x) is generally a decreasing function of x and the asymptotic behav-iour of the reliability function of the MOGETE distribution is R(0) = 1; and R(∞) = 0.
3.2. Hazard rate functionThe hazard rate function (hrf) gives the probability of failure for a system that has survived up-to time x. It is mathematically expressed as h(x) = f (x)∕R(x). The hazard rate function of the MOGETE distribution is given by
Theorem 3.1 The shape of the hrf of the MOGETE distribution is constant (a special case of the ETE distribution when � = 1), decreasing when 𝛼 < 1 and increasing when 𝛼 > 1.
Proof The first derivative of the hrf h′ in Equation (9) is given by
f � = −𝛼𝛽
2(1 − e−𝜆)2e−𝛽(1−e−𝜆)x
(1 − (1 − 𝛼)e−𝛽(1−e−𝜆)x)2
−2𝛼(1 − 𝛼)𝛽2(1 − e−𝜆)2e−2𝛽(1−e
−𝜆)x
(1 − (1 − 𝛼)e−𝛽(1−e−𝜆)x)3
< 0.
x0 =log(𝛼 − 1)
𝛽(1 − e−𝜆); ∀ 𝛼 > 2,
(8)R(x) =𝛼e−𝛽(1−e
−𝜆)x
1 − (1 − 𝛼)e−𝛽(1−e−𝜆)x; 0 ≤ x < ∞, 𝛼 > 0, 𝛽, 𝜆 > 0
(9)h(x) =𝛽(1 − e−𝜆)
1 − (1 − 𝛼)e−𝛽(1−e−𝜆)x; 0 ≤ x < ∞, 𝛼 > 0, 𝛽, 𝜆 > 0.
h� = −(1 − �)�2(1 − e−�)2e−�(1−e
−�)x
(1 − (1 − �)e−�(1−e−�)x)2
.
Page 5 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
It is easy to see that h� = 0 has no unique root; h� < 0, ∀ 𝛼 < 1 (i.e. the hrf is decreasing) and h� > 0, ∀𝛼 > 1 (i.e. the hrf is increasing) and when � = 1, h� = 0 (i.e. the hrf is constant). ✷
The asymptotic behaviour of the hrf of the MOGETE distribution is h(0) = �(1 − e−�)∕� and h(∞) = �(1 − e−�).
3.3. The mean residual life time
Theorem 3.2 The remaining lifetime of a system that has survived up-to time t is random, as a result the failure time cannot be determined. The expected value of the random failure times is referred to as the mean residual lifetime denoted by M(t). M(t) only exists for F(t) > 0 and it is mathematical expressed as
The mean residual lifetime of the MOGETE distribution is given by
Proof
Substituting y = �(1 − e−�)x into Equation (11), we have
✷
4. Some statistical properties of the MOGETE distributionApplication of any distribution can only be possible if its basic distributional properties are available. In this section, we present explicit derivations of some important distributional properties of the MOGETE distribution.
4.1. The pth quantile function of the MOGETE distributionThe pth quantile function of the MOGETE distribution is given by
Random variables can be simulated from the MOGETE distribution through the method of the in-version of cdf by simply substituting p in Equation (12) with a U(0, 1) variates. Also, the median of the
M(t) = E(X − t|X > t) =1
R(t)
∞
∫t
R(x)dx
(10)M(t) =
�
R(t)�(1 − e−�)
∞∑j=0
(−1)2j(1 − �)je−[�(1−e
−�)t](j+1)
j + 1.
(11)M(t) =1
R(t)
∞
∫t
�e−�(1−e−�)x[1 − (1 − �)e−�(1−e
−�)x]−1dx
M(t) =�
R(t)�(1 − e−�)
∞
∫�(1−e−�)t
e−y[1 − (1 − �)e−y
]−1dy
=�
R(t)�(1 − e−�)
∞
∫�(1−e−�)t
e−y∞∑j=0
(−1)j
(j + 1 − 1
j
)(−1)j(1 − �)je−yjdy
=�
R(t)�(1 − e−�)
∞∑j=0
(−1)2j(1 − �)j
∞
∫�(1−e−�)t
e−y(j+1)dy
=�
R(t)�(1 − e−�)
∞∑j=0
(−1)2j(1 − �)je−[�(1−e
−�)t](j+1)
j + 1.
(12)�(p) = −1
�(1 − e−�)log
[1 − p
1 − p(1 − �)
]
Page 6 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
MOGETE distribution could be obtained from Equation (12) by setting p = 1∕2. The Bowley skewness denoted by B due to Bowley (1901–1920) and Moors kurtosis denoted by M due to Moors (1986) depends on the quantile function.
The Bowley skewness is given by
and, the Moors kurtosis is given by
Figure 1 illustrates the variability of both the Bowley skewness and Moors kurtosis on �(shape param-eter) for the MOGETE distribution. The Bowley skewness and Moors kurtosis shrinks with increasing value of �.
4.2. The kth crude moment of the MOGETE distribution
Theorem 4.1 If the kth crude moment of any random variable X exists then other essential charac-teristics of the distribution could be derived from it, such as the mean, variance, coefficient of varia-tion, skewness and kurtosis statistics. The kth crude moment of any continuous random variable X is generally given by E(Xk) = ∫∞
−∞xkf (x)dx. Hence, it follows that the kth crude moment of the MOGETE
distribution is given by
Proof
B =�(3∕4) + �(1∕4) − 2�(2∕4)
�(3∕4) − �(1∕4),
M =�(3∕8) − �(1∕8) + �(7∕8) − �(5∕8)
�(6∕8) − �(2∕8).
E(Xk) =�Γ(k + 1)
[�(1 − e−�)
]k∞∑j=0
(−1)2j(1 − �)j
(j + 1)k.
(13)
E(Xk) =
∞
∫0
xk��(1 − e−�)e−�(1−e−�)x[1 − (1 − �)e−�(1−e
−�)x]−2dx
= ��(1 − e−�)
∞
∫0
xke−�(1−e−�)x[1 − (1 − �)e−�(1−e
−�)x]−2dx
Figure 1. Plots of the Bowley skewness (left panel) and Moors kurtosis (right panel) of the MOGETE distribution.
Page 7 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Substituting y = �(1 − e−�)x into Equation (13) gives
Substituting z = y(j + 1) into Equation (14) gives
✷
The mean is the first-order crude moment of the distribution and could be obtained by evaluating Equation (15) at k = 1 as
While evaluating Equation (15) at k = 2 gives the second-order crude moment of the MOGETE distri-bution as
The variance V(X) could be obtained by substituting E(X) and E(X2) into the following expression V(X) = E(X2) − {E(X)}2. Hence, the variance of the MOGETE distribution is given by
Setting E(Xk) = ��
k the coefficient of variation (CV), skewness (�1) and kurtosis (�2) statistics of the MOGETE distribution could be obtained as follows
(14)
=�[
�(1 − e−�)]k
∞
∫0
yke−y[1 − (1 − �)e−y
]−2dy
=�[
�(1 − e−�)]k
∞
∫0
yke−y∞∑j=0
(−1)j
(j + 2 − 1
j
)(−1)j(1 − �)je−yjdy
=�[
�(1 − e−�)]k
∞∑j=0
(−1)2j(j + 1)(1 − �)j
∞
∫0
yke−y(j+1)dy
(15)
=�[
�(1 − e−�)]k
∞∑j=0
(−1)2j(1 − �)j
(j + 1)k
∞
∫0
zke−zdz
=�Γ(k + 1)
[�(1 − e−�)
]k∞∑j=0
(−1)2j(1 − �)j
(j + 1)k.
E(X) =�
�(1 − e−�)
∞∑j=0
(−1)2j(1 − �)j
j + 1.
E(X2) =2�[
�(1 − e−�)]2
∞∑j=0
(−1)2j(1 − �)j
(j + 1)2.
V(X) =2�[
�(1 − e−�)]2
∞∑j=0
(−1)2j(1 − �)j
(j + 1)2−
[�
�(1 − e−�)
∞∑j=0
(−1)2j(1 − �)j
j + 1
]2
CV =
√√√√ ��
2
��21
− 1
�1 =��
3 − 3��
2��
1 + 2��31
(��
2 − ��21 )
3
2
�2 =��
4 − 4��
3��
1 + 6��
2��21 − 3��4
1
(��
2 − ��21 )
2
Page 8 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
4.3. The kth central moment of the MOGETE distribution
Theorem 4.2 The kth central moment of a continuous random variable X is given by E((X − �)k) = ∫∞
−∞(x − �)kf (x)dx. Hence, the kth central moment of the MOGETE distribution is given by
Proof
Substituting y = �(1 − e−�)x into Equation (16) gives
Substituting z = y(j + 1) into Equation (17) gives
✷
4.4. The moment generating function of the MOGETE distributionRecently, a lot of advancement both in theory and application has been achieved in statistics and probability through the moment generating function (mgf) of a random variable X. The usefulness of the mgf has been found to surpass the very trivial derivation of distributional order moments. For
E[(X − �)k] =�[
�(1 − e−�)]k
∞∑i=0
∞∑j=0
(−1)i+2jΓ(k + 1)
[��(1 − e−�)
]i(1 − �)j(j + 2)Γ(k − i + 1)
Γ(i + 1)Γ(k − i + 1)(j + 1)k−i+1.
(16)
E[(X − �)k] =
∞
∫0
(x − �)k��(1 − e−�)e−�(1−e−�)x[1 − (1 − �)e−�(1−e
−�)x]−2dx
= ��(1 − e−�)
∞
∫0
(x − �)ke−�(1−e−�)x[1 − (1 − �)e−�(1−e
−�)x]−2dx
(17)
E[(X − �)k] =�[
�(1 − e−�)]k
∞
∫0
e−y[y − ��(1 − e−�)
]k
⋅
[1 − (1 − �)e−y
]−2dy
=�[
�(1 − e−�)]k
∞
∫0
e−y∞∑i=0
(k
i
)(−1)i
[��(1 − e−�)
]iyk−i
⋅
∞∑j=0
(−1)j
(j + 2 − 1
j
)(−1)j(1 − �)je−yjdy
=�[
�(1 − e−�)]k
∞∑i=0
∞∑j=0
(−1)i+2j[��(1 − e−�)
]i(1 − �)j(j + 2)
(k
i
)
⋅
∞
∫0
yk−ie−y(j+1)dy
E[(X − �)k] =�[
�(1 − e−�)]k
∞∑i=0
∞∑j=0
(−1)i+2j
(k
i
)[��(1 − e−�)
]i(1 − �)j(j + 2)
(j + 1)k−i+1
⋅
∞
∫0
zk−ie−zdz
=�[
�(1 − e−�)]k
∞∑i=0
∞∑j=0
(−1)i+2j
⋅
Γ(k + 1)[��(1 − e−�)
]i(1 − �)j(j + 2)Γ(k − i + 1)
Γ(i + 1)Γ(k − i + 1)(j + 1)k−i+1
Page 9 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
instance; Villa and Escobar (2006) obtained mixture distributions with mgf, Meintanis (2010) used the mgf for testing skew normality, McLeish (2014) performed simulation of random variables using the mgf and the saddle point approximation, von Waldenfels (1987) gave a proof of an algebraic central limit theorem using the mgf, and Inlow (2010) also, proved the Lindeberg-Lévy’s central limit theorem with the mgf. The moment generating function is generally defined by
It follows from Equations (15) and (18) that the mgf of the MOGETE distribution is given by
4.5. Rényi entropy of the MOGETE distributionThe Rényi entropy denoted by H
�(x) is used to quantify the uncertainty of variation in a random vari-
able X. The limiting value of H�(x) as � → 1 is the Shannon entropy. Song (2001) compared tails and
shapes behaviour of some standard probability distributions using the Rényi entropy. The Rényi en-tropy measure is generally given by
Theorem 4.3 If X follows the MOGETE distribution, then its Rényi entropy measure is given by
Proof
Substituting y = �(1 − e−�)x into Equation (19) gives
✷
(18)MX(t) = E(etx) = E
[∞∑k=0
(tx)k
k!
]=
∞∑k=0
tk
k!E(xk).
MX(t) =
∞∑j=0
∞∑k=0
(−1)2jtk�Γ(k + 1)
k![�(1 − e−�)
]k(1 − �)j
(j + 1)k.
H�(x) = lim
n→∞(I
�(fn) − log(n)) =
1
1 − �log ∫ f �(x)dx =
1
1 − �log(I
�).
H�(x) =
1
1 − �log
([��
(1 − e−�
)]��(1 − e−�)
∞∑j=0
(−1)2jΓ(j + 2�)
Γ(j + 1)Γ(2�)
(1 − �)j
� + j
).
(19)
I�=
∞
∫0
[��
(1 − e−�
)e−�(1−e
−�)x(1 − (1 − �)e−�(1−e
−�)x)−2
]�dx
=[��
(1 − e−�
)]�∞
∫0
e−��(1−e−�)x
(1 − (1 − �)e−�(1−e
−�)x)−2�
dx
I�=
[��
(1 − e−�
)]��(1 − e−�)
∞
∫0
e−�y(1 − (1 − �)e−y
)−2�dy
=
[��
(1 − e−�
)]��(1 − e−�)
∞
∫0
e−�y∞∑j=0
(−1)j
(j + 2� − 1
j
)(−1)j(1 − �)je−yjdy
=
[��
(1 − e−�
)]��(1 − e−�)
∞∑j=0
(−1)2j
(j + 2� − 1
j
)(1 − �)j
∞
∫0
e−y(�+j)dy
=
[��
(1 − e−�
)]��(1 − e−�)
∞∑j=0
(−1)2j
(j + 2� − 1
j
)(1 − �)j
� + j
H�(x) =
1
1 − �log
([��
(1 − e−�
)]��(1 − e−�)
∞∑j=0
(−1)2jΓ(j + 2�)
Γ(j + 1)Γ(2�)
(1 − �)j
� + j
)
Page 10 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
4.6. Order statistics of the MOGETE random variableThe distribution of the rth order statistics denoted by fX
(r)(x) of an n sized random sample
X1,X2,X3,… ,Xn is generally given by
The density of the rth order statistics of the MOGETE distribution could be obtained by substituting Equations (6) and (7) into Equation (20) as
The density of the rth smallest order statistics of the MOGETE distribution is given by
The density of the rth largest order statistics of the MOGETE distribution is given by
5. EstimationHere, we propose to estimate the parameters of the MOGETE distribution by the method of Maximum likelihood estimation.
5.1. Maximum likelihood estimation methodSuppose the random sample x1, x2, x3,… , xn of size n is drawn from a probability distribution with pdf f(x) then the maximum likelihood estimates (mle) of its parameters could be obtained as follows:
The likelihood () equation is given by
and the log-likelihood (�) equation is given by
then; taking the partial derivatives of Equation (22) w.r.t to �; � and � and equating to zero gives:
(20)fX(r)(x) =
n!
(r − 1)!(n − r)!(Fx(x))
r−1(1 − Fx(x))n−r fx(x).
fX(k)(x) =
n!��(1 − e−�
)e−�(1−e
−�)x[1 − (1 − �)e−�(1−e
−�)x]−2
(k − 1)!(n − k)!
⋅
[1 − e−�(1−e
−�)x
1 − (1 − �)e−�(1−e−�)x
]k−1[�e−�(1−e
−�)x
1 − (1 − �)e−�(1−e−�)x
]n−k
fX(1)(x) = n!��
(1 − e−�
)e−�(1−e
−�)x
⋅
[1 − (1 − �)e−�(1−e
−�)x]−2[ �e−�(1−e
−�)x
1 − (1 − �)e−�(1−e−�)x
]n−1
fX(n)(x) = n!��
(1 − e−�
)e−�(1−e
−�)x
⋅
[1 − (1 − �)e−�(1−e
−�)x]−2[ 1 − e−�(1−e
−�)x
1 − (1 − �)e−�(1−e−�)x
]n−1
(21)
=
n�i=1
��(1 − e−�)e−�(1−e−�)xi
�1 − (1 − �)e−�(1−e
−�)xi
�−2
=���(1 − e−�)
�ne−�(1−e−�)
n∑i=1
xin�i=1
�1 − (1 − �)e−�(1−e
−�)xi
�−2
(22)� = n log(��(1 − e−�)
)− �(1 − e−�)
n∑i=1
xi − 2
n∑i=1
log{1 − (1 − �)e−�(1−e
−�)xi
}
Page 11 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
The analytical solution to the system of nonlinear equations in Equations (23), (24) and (25) does not exist thus, we require some nonlinear numerical optimization methods such as the Newton Raphson technique to solve the equations.
Let � = (��, 𝛽, ��)�. Under some standard regularity conditions, √n(� −�) is asymptotically mul-
tivariate normal distributed �3(�, �−1n (�)), where �n(�) is the expected information matrix defined
by �(�2�(�)∕�����). The asymptotic behaviour of the expected information matrix can be ap-
proximated by the observed information matrix, denoted by �n(�). Generally speaking, the diagonal elements of �−1n (�) gives the variance of (�) while the off-diagonal elements is the covariances. The observed information matrix of the MOGETE distribution is expressed as
where the corresponding elements are:
(23)��
��=n
�− 2
n∑i=1
e−�(1−e−�)xi
1 − (1 − �)e−�(1−e−�)xi
= 0
(24)��
��=n
�− (1 − e−�)
n∑i=1
xi − 2
n∑i=1
(1 − �)(1 − e−�)xie−�(1−e−�)xi
1 − (1 − �)e−�(1−e−�)xi
= 0
(25)��
��=
ne−�
1 − e−�− �e−�
n∑i=1
xi − 2
n∑i=1
(1 − �)�xie−�−�(1−e−�)xi
1 − (1 − �)e−�(1−e−�)xi
= 0
�n(�) =
⎛⎜⎜⎜⎝
𝜕2�(�)
𝜕𝛼2
𝜕2�(�)
𝜕𝛼𝜕𝛽
𝜕2�(�)
𝜕𝛼𝜕𝜆
𝜕2�(�)
𝜕𝛽𝜕𝛼
𝜕2�(�)
𝜕𝛽2
𝜕2�(�)
𝜕𝛽𝜕𝜆
𝜕2�(�)
𝜕𝜆𝜕𝛼
𝜕2�(�)
𝜕𝜆𝜕𝛽
𝜕2�(�)
𝜕𝜆2
⎞⎟⎟⎟⎠.
�2�
��2= −
n
�2+ 2
n∑i=1
e−2�(1−e−�)x
i
[1 − (1 − �)e−�(1−e−�)x
i ]2,
�2�
��2= −
n
�2+ 2
n∑i=1
(1 − �)(1 − e−�)2x2ie−�(1−e−�)x
i
1 − (1 − �)e−�(1−e−�)x
i
+ 2
n∑i=1
(1 − �)2(1 − e−�)2x2ie−2�(1−e−�)x
i
(1 − (1 − �)e−�(1−e−�)x
i )2,
�2�
��2= −
n[e−�(1 − e−�) + e−2�]
(1 − e−�)2+ �e
−�
n∑i=1
xi
− 2
n∑i=1
[(1 − �)�x
ie−�−�(1−e−�)x
i (−1 − �xie−�)
1 − (1 − �)e−�(1−e−�)x
i
−(1 − �)2�2x2
ie−2(�+(1−e−�)x
i)
(1 − (1 − �)e−�(1−e−�)x
i )2
],
�2�
����= 2
n∑i=1
(1 − e−�)xie−�(1−e−�)x
i
1 − (1 − �)e−�(1−e−�)x
i
+ 2
n∑i=1
(1 − �)(1 − e−�)xie−2�(1−e−�)x
i
[1 − (1 − �)e−�(1−e−�)x
i ]2,
�2�
����= 2
n∑i=1
�xie−�−�(1−e−�)x
i
1 − (1 − �)e−�(1−e−�)x
i
+ 2
n∑i=1
(1 − �)�xie−�−2�(1−e−�)x
i
[1 − (1 − �)e−�(1−e−�)x
i ]2,
Page 12 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
and
Given that √n(� −� ∼)�
3(�, �−1
n(�)), we can perform statistical inference for functions of �. For
instance, the approximate 100(1 − �)% two-sided confidence interval of the model parameters � could be calculated as:
where �−1�2𝓁(�)
���
(⋅), are the diagonal entries of the observed information matrix, and Z �
2
is the upper
�∕2th percentile of the standard normal distribution.
5.1.1. Simulation studyOne major problem of extended probability distributions is parameter estimation. In this section, we present a Monte Carlo simulation study to evaluate the performance of the mle method in estimat-ing the parameters of the distribution by drawing different samples (n = 50, 100, … , 300) from the MOGETE distribution with selected parameter values. Estimation of the parameters was carried out with the simulated random variables through the mle method to investigate the stability of the parameters and sample size effect on the estimates via bias, standard error (se), and mean square error (mse). Application of the following algorithm in (Statistical software) provides us with the results in Table 1.
5.1.2. Algorithm
(i) Simulate ui∼ Uniform(0, 1), for i = 1, 2, 3,… ,n(50, 100,… , 300);
(ii) Set xi= F−1(u
i), where F−1(⋅) is the quantile function in Equation (17) evaluated at U
i for some
parameter values (see Table 1) and X ~ MOGETE distribution;
(iii) Using x and the nlm function under the stats package in , calculate the mle estimates of the parameters of the MOGETE distribution;
(iv) Repeat steps (i–iii) in 5,000 (N) iterations;
(v) For each n and parameter, compute the mean (parameter estimate), standard deviation (standard error), bias and mse of the sequence of 5,000 parameter estimates.
�2�
����= −e−�
n∑i=1
xi − 2
n∑i=1
(1 − �)xie−�−�(1−e−�)xi
1 − (1 − �)e−�(1−e−�)xi
+ 2
n∑i=1
(1 − �)�x2i e−�−�(1−e−�)xi
1 − (1 − �)e−�(1−e−�)xi
+ 2
n∑i=1
(1 − �)2�(1 − e−�)x2i e−�−2�(1−e−�)xi
[1 − (1 − �)e−�(1−e−�)xi ]2
� ± Z �
2
√�−1�2𝓁(�)
���
(⋅),
Page 13 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Tabl
e 1.
Sim
ulat
ion
resu
ltsn
𝜶 𝜷
𝝀se
𝜶se
𝜷se
𝝀bias 𝜶
bias
𝜷bias 𝝀
mse
𝜶mse
𝜷mse
𝝀
�=5.000,�=5.000,�=5.000
506.
1859
85.
2149
95.
0050
63.
4368
10.
9572
70.
0335
31.
1859
80.
2149
90.
0050
613
.215
860.
9624
00.
0011
5
100
5.56
314
5.10
831
5.00
252
1.99
627
0.65
025
0.02
124
0.56
314
0.10
831
0.00
252
4.30
142
0.43
447
0.00
046
150
5.33
840
5.07
000
5.00
156
1.53
315
0.53
264
0.01
763
0.33
840
0.07
000
0.00
156
2.46
458
0.28
855
0.00
031
200
5.24
952
5.04
803
5.00
105
1.32
411
0.46
110
0.01
526
0.24
952
0.04
803
0.00
105
1.81
516
0.21
488
0.00
023
250
5.19
484
5.03
778
5.00
101
1.14
615
0.41
452
0.01
388
0.19
484
0.03
778
0.00
101
1.35
136
0.17
322
0.00
019
300
5.15
584
5.03
225
5.00
086
1.02
739
0.37
190
0.01
250
0.15
584
0.03
225
0.00
086
1.07
961
0.13
933
0.00
016
�=2.500,�=3.500,�=4.500
503.
0341
83.
6819
04.
5053
11.
6366
10.
7981
20.
0299
20.
5341
80.
1819
00.
0053
12.
9633
20.
6699
50.
0009
2
100
2.72
910
3.58
448
4.50
254
0.95
544
0.54
492
0.02
080
0.22
910
0.08
448
0.00
254
0.96
518
0.30
402
0.00
044
150
2.66
079
3.56
598
4.50
229
0.75
167
0.44
686
0.01
716
0.16
079
0.06
598
0.00
229
0.59
075
0.20
400
0.00
030
200
2.61
593
3.54
175
4.50
137
0.62
960
0.37
883
0.01
472
0.11
593
0.04
175
0.00
137
0.40
976
0.14
523
0.00
022
250
2.58
400
3.52
840
4.50
101
0.55
709
0.33
495
0.01
304
0.08
400
0.02
840
0.00
101
0.31
735
0.11
297
0.00
017
300
2.57
327
3.52
693
4.50
100
0.51
057
0.31
197
0.01
214
0.07
327
0.02
693
0.00
100
0.26
600
0.09
803
0.00
015
�=8.900,�=2.300,�=0.500
5010
.974
842.
2946
90.
5284
66.
2933
70.
0331
50.
1129
42.
0748
4−0.00531
0.02
846
43.9
0350
0.00
113
0.01
356
100
9.89
336
2.29
899
0.51
355
3.78
194
0.02
139
0.07
380
0.99
336
−0.00101
0.01
355
15.2
8695
0.00
046
0.00
563
150
9.51
749
2.29
901
0.50
832
2.87
953
0.01
755
0.05
901
0.61
749
−0.00099
0.00
832
8.67
133
0.00
031
0.00
355
200
9.36
451
2.29
955
0.50
658
2.46
095
0.01
492
0.05
059
0.46
451
−0.00045
0.00
658
6.27
081
0.00
022
0.00
260
250
9.28
069
2.29
999
0.50
541
2.15
863
0.01
332
0.04
502
0.38
069
−0.00001
0.00
541
4.80
369
0.00
018
0.00
206
300
9.17
078
2.30
029
0.50
372
1.85
221
0.01
184
0.03
961
0.27
078
0.00
029
0.00
372
3.50
334
0.00
014
0.00
158
�=0.500,�=0.500,�=0.500
500.
6322
70.
5308
80.
5241
50.
3567
40.
1184
10.
0918
30.
1322
70.
0308
80.
0241
50.
1447
40.
0149
70.
0090
1
100
0.56
720
0.51
692
0.51
308
0.22
019
0.08
190
0.06
312
0.06
720
0.01
692
0.01
308
0.05
299
0.00
699
0.00
415
150
0.54
571
0.51
280
0.50
988
0.16
805
0.06
649
0.05
124
0.04
571
0.01
280
0.00
988
0.03
032
0.00
458
0.00
272
200
0.52
970
0.50
844
0.50
651
0.14
004
0.05
660
0.04
363
0.02
970
0.00
844
0.00
651
0.02
049
0.00
327
0.00
195
250
0.52
688
0.50
708
0.50
546
0.12
643
0.05
079
0.03
914
0.02
688
0.00
708
0.00
546
0.01
670
0.00
263
0.00
156
300
0.52
068
0.50
604
0.50
465
0.11
498
0.04
709
0.03
630
0.02
068
0.00
604
0.00
465
0.01
365
0.00
225
0.00
134
�=20.000,�=20.000,�=1.000
5026
.423
5019
.994
021.
1177
118
.116
900.
1376
50.
4397
76.
4235
0−0.00598
0.11
771
369.
4178
70.
0189
80.
2072
1
100
22.3
5379
19.9
9680
1.04
063
9.24
222
0.02
020
0.19
696
2.35
379
−0.00320
0.04
063
90.9
4194
0.00
042
0.04
044
150
21.8
4860
19.9
9840
1.03
121
7.46
007
0.01
590
0.15
816
1.84
860
−0.00160
0.03
121
59.0
5884
0.00
026
0.02
598
200
21.1
9487
19.9
9851
1.02
011
5.89
144
0.01
853
0.13
151
1.19
487
−0.00149
0.02
011
36.1
2989
0.00
035
0.01
770
250
20.9
0419
19.9
9874
1.01
528
5.10
803
0.01
032
0.11
398
0.90
419
−0.00126
0.01
528
26.9
0432
0.00
011
0.01
322
300
20.8
5828
19.9
9939
1.01
474
4.66
581
0.02
018
0.10
550
0.85
828
−0.00061
0.01
474
22.5
0210
0.00
041
0.01
135
Page 14 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Whatever parameter values we chose in Table 1, the mle estimates ( � = 1∕Nn∑i=1
�i) approximates
to the actual value as n becomes large and the standard error �se
�=
�∑N
i=1(�
i−
�)2∕(N − 1)
�,
bias (bias = �i−�; i = 1,… ,n) and mse (mse = 1∕N
∑N
i=1[�
i−�]2) decreases with increasing
n, where � = (�, �, �)�. These results suggest that the mle method does well in estimating the pa-rameters of the MOGETE distribution.
6. ApplicationsThis section illustrates the applicability and flexibility of the MOGETE distribution with two real data-sets. The goodness of fit of the new lifetime distribution would be assessed by a comparison of its performance in modelling the real data-sets with that of its competing sub-model (ETE distribution) and the following three-parameter distributions:
• Exponentiated Burr XII
• Exponentiated Fréchet
based on the Akaike information criterion (AIC) statistic, Akaike (1981),
the AIC with a correction statistic (AICc), Sugiura (1978),
where �, k, and n corresponds to the estimate of the model maximized/minimized log-likelihood function, number of model parameters and sample size, respectively. The Chen and Balakrishnan (1995) W⋆ and A⋆ goodness of fit measures were also considered. See Oluyede, Foya, Warahena-Liyanage, and Huang (2016) for detail on the computational steps of the W⋆ and A⋆ statistics. The distribution with the smallest goodness-of-fit measure is the best. Table 2 gives the waiting times in minutes of 100 bank customers in a queue before service. The data-set was first published in Ghitany, Atieh, and Nadarajah (2008). Merovci and Elbatal (2013) and Bhati, Malik, and Vaman (2015) have also fitted the data to different distributions. The results we obtained from the data fitting are tabu-lated in Table 3.
f (x) = 𝛼kcxc−1(1 + xc)−k−1[1 − (1 + xc)−k]𝛼−1, 𝛼, c, k > 0, and
f (x) =𝛼𝛽
s
(x
s
)−𝛽−1
e−𝛼(x
s)−𝛽
, 𝛼, 𝛽, s > 0,
AIC = −2� + 2k
AICc = AIC +2k(k + 1)
n − k − 1
Table 2. 100 bank customers waiting times (min) before service0.8 0.8 1.3 1.5 1.8 1.9 1.9 2.1 2.6 2.7
2.9 3.1 3.2 3.3 3.5 3.6 4.0 4.1 4.2 4.2
4.3 4.3 4.4 4.4 4.6 4.7 4.7 4.8 4.9 4.9
5.0 5.3 5.5 5.7 5.7 6.1 6.2 6.2 6.2 6.3
6.7 6.9 7.1 7.1 7.1 7.1 7.4 7.6 7.7 8.0
8.2 8.6 8.6 8.6 8.8 8.8 8.9 8.9 9.5 9.6
9.7 9.8 10.7 10.9 11.0 11.0 11.1 11.2 11.2 11.5
11.9 12.4 12.5 12.9 13.0 13.1 13.3 13.6 13.7 13.9
14.1 15.4 15.4 17.3 17.3 18.1 18.2 18.4 18.9 19.0
19.9 20.6 21.3 21.4 21.9 23.0 27.0 31.6 33.1 38.5
Page 15 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
The variance-covariance matrix of the MOGETE distribution under the fitted 100 bank customers waiting times data is given by
The second example is on the annual maximum daily precipitation in millimetre that was recorded in Basan, Korea, from 1904 to 2011. The data are presented in Table 4. The data-set has been ana-lysed by Jeong, Murshed, Am Seo, and Park (2014) and was recently reported in Mansoor et al. (2016). Results from the data fitting to the distributions are presented in Table 5.
�−1n (�) =
⎛⎜⎜⎝
1.83309877 0.2559214 0.00352501
0.25592137 85.9418670 −4.10769366
0.00352501 −4.1076937 0.19651703
⎞⎟⎟⎠
Table 3. Results from modelling the 100 bank customers waiting times dataModels Estimatess s.e −� AIC AICc W
⋆A⋆
MOGETE
�� 4.11562602 1.353920
𝛽 2.10379157 9.270484 320.7120 647.4241 647.6741 0.0008333 0.6351718
�� 0.09588135 0.443302
ETE
𝛽 0.2428209 5.284512 329.0209 662.0418 662.1655 27.8976100 4.2290630
�� 0.5394967 15.56389
Exp Fréchet
�� 4.200080 1.978491e+03
𝛽 1.162912 7.995092e-02 334.3810 674.7620 675.0120 0.3832246 2.504610
s 1.461971 5.921992e+02
Exp Burr XII
�� 39.882445 39.3966644
k 2.829186 1.2939812 327.5301 661.0601 661.3101 0.2297712 1.561101
c 0.595938 0.2128391
Table 4. Rainfall data24.8 140.9 54.1 153.5 47.9 165.5 68.5 153.1 254.7 175.3 87.6 150.6
147.9 354.7 128.5 150.4 119.2 69.7 185.1 153.4 121.7 99.3 126.9 150.1
149.1 143 125.2 97.2 179.3 125.8 101 89.8 54.6 283.9 94.3 165.4
48.3 69.2 147.1 114.2 159.4 114.9 58.5 76.6 20.7 107.1 244.5 126
122.2 219.9 153.2 145.3 101.9 135.3 103.1 74.7 174 126 144.9 226.3
96.2 149.3 122.3 164.8 188.6 273.2 61.2 84.3 130.5 96.2 155.8 194.6
92 131 137 106.8 131.6 268.2 124.5 147.8 294.6 101.6 103.1 247.5
140.2 153.3 91.8 79.4 149.2 168.6 127.7 332.8 261.6 122.9 273.4 178
177 108.5 115 241 76 127.5 190 259.5 301.5
Page 16 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
The variance–covariance matrix of the MOGETE distribution under the fitted Rainfall data is given by
7. Discussion of resultsThe density plots in Figure 2 (left panel) depict some monotonic decreasing function of x for � ≤ 2 and for α > 2 the distribution is unimodal, while the cdf plot (right panel) shows some monotonic increasing curves for all 𝛼 < 1. The plots in Figure 2 indicate that the reliability function (left panel) is a monotonically decreasing function of x for all � while the hazard rate function (right panel) could increasing (if 𝛼 < 1), decreasing (if 𝛼 < 1), or constant (if � = 1), these characteristics make it more reasonable for analysing complex lifetime data-sets. The results in Tables 3 and 5 show that the MOGETE distribution with smaller minimized log-likelihood value and smaller information statistics provides better fit to the data-sets than the ETE and the other competing distributions. Also, the P-P plots in Figures 4 and 5 does not raise any alarm against the suggestion of the AIC, AICc, W* and A⋆ statistics.
8. ConclusionsThis article introduces a new lifetime distribution—the (MOGETE) distribution. The new distribution generalizes the ETE distribution and has the ETE distribution as a sub-model. We have given explicit mathematical expressions for some of its basic statistical properties such as the probability density function, cumulative density function, kth raw moment,kth central moment, mean, variance, coef-ficient of variation, skewness, kurtosis, moment generating function, pth quantile function, the rth order statistics and the Rényi’s entropy measure. Also, some of its reliability characteristics like the
�−1n (�) =
⎛⎜⎜⎝
193.60658478 0.2749396 0.03408826
0.27493958 0.4427812 −0.11510233
0.03408826 −0.1151023 0.02999145
⎞⎟⎟⎠
Table 5. Results from modelling the rainfall dataModels Estimates s.e −� AIC AICc W
⋆A⋆
MOGETE
�� 35.58880229 13.9142583
𝛽 0.33061261 0.6654180 583.7588 1173.518 1173.755 0.0007939 1.2923280
�� 0.08257185 0.1731804
ETE
𝛽 0.01535297 0.01651167 627.2663 1258.533 1258.650 104.3609000 15.7744200
�� 0.59875101 0.8855356
Exp Fréchet
�� 14.21029 227.2744212
𝛽 1.64995 0.1031434 608.5969 1223.194 1223.432 0.7123182 4.1213390
s 19.99516 193.8275346
Exp Burr XII
�� 591.1682852 181.9098298
k 0.9348673 0.4196369 612.2200 1230.440 1230.678 0.5777131 3.3452710
c 1.4783094 0.6605398
Page 17 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
Figure 4. Probability-Probability (P-P) plots of the fitted distributions with the waiting times data.
Figure 3. Possible shapes of the survival/reliability function F(x) (left) and hazard rate function h(x) (right) of the MOGETE distribution for fixed parameter values of � and � and selected values of � parameter.
Figure 2. Possible shapes of the probability density function f (x) (left) and cumulative distribution function F(x) (right) of the MOGETE distribution for fixed parameter values of � and � and selected values of � parameter.
Page 18 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
reliability function, hazard rate function and the mean residual life time was given. Estimation of the model parameters was approached through the method of maximum likelihood estimation. The applicability, flexibility and robustness of the new lifetime distribution was demonstrated with the 100 bank customers waiting times data and 105 Rainfall data, and the results obtained show that the MOGETE distribution provides a more reasonable fit than the ETE, Exp Fréchet and Exp Burr XII distributions. We hope that the MOGETE distribution would receive a high rate of application, par-ticularly, because of its hazard rate characteristics.
Figure 5. Probability-Probability (P-P) plots of the fitted distributions with the Rainfall data.
FundingThis work was supported by the University of Manchester, UK.
Author detailsIdika E. Okorie1
E-mail: [email protected] ID: http://orcid.org/0000-0001-7770-3036Anthony C. Akpanta2
E-mail: [email protected] ID: http://orcid.org/0000-0003-4178-9370Johnson Ohakwe3
E-mail: [email protected] ID: http://orcid.org/0000-0001-5193-86541 School of Mathematics, University of Manchester,
Manchester M13 9PL, UK.2 Department of Statistics, Abia State University, Uturu, Abia
State, Nigeria.3 Faculty of Sciences, Department of Mathematics & Statistics,
Federal University Otuoke, P.M.B 126 Yenagoa, Bayelsa, Bayelsa State, Nigeria.
Citation informationCite this article as: Marshall-Olkin generalized Erlang-truncated exponential distribution: Properties and applications, Idika E. Okorie, Anthony C. Akpanta & Johnson Ohakwe, Cogent Mathematics (2017), 4: 1285093.
ReferencesAkaike, H. (1981). Likelihood of a model and information
criteria. Journal of Econometrics. 16, 3–14.Alizadeh, M., Tahir, M. H., Cordeiro, G. M., Mansoor, M., Zubair,
M., & Hamedani, G. G. (2015). The Kumaraswamy Marshal-Olkin family of distributions. Journal of the Egyptian Mathematical Society, 23, 546–557.
Al-Saiari, A. Y., Baharith, L. A., & Mousa, S. A. (2014). Marshall-Olkin extended Burr type XII distribution. International Journal of Statistics and Probability, 3, 78.
Barreto-Souza, W., Lemonte, A. J., & Cordeiro, G. M. (2013). General results for the Marshall and Olkin’s family of distributions. Anais da Academia Brasileira de Cincias, 85, 3–21.
Bhati, D., Malik, M. A., & Vaman, H. J. (2015). Lindley-Exponential distribution: Properties and applications. Metron, 73, 335–357.
Bowley, A. L. (1901–1920). Elements of statistics (4th ed.). New York, NY: Charles Scribner.
Chen, G., & Balakrishnan, N. (1995). A general purpose approximate goodness-of-fit test. Journal of Quality Technology, 27, 154–161.
El-Alosey, A. R. (2007). Random sum of new type Of mixture Of distribution. International Journal Of Statistics and Systems, 2, 49–57.
Ghitany, M. E. (2005). Marshall--Olkin extended Pareto and its application. International Journal of Applied Mathematics, 18, 17–32.
Ghitany, M. E., Al-Awadhi, F. A., & Alkhalfan, L. A. (2007). Marshall–-Olkin extended Lomax distribution and its application to censored data. Communications in Statistics Theory and Methods, 36, 1855–1866.
Ghitany, M. E., Al-Hussaini, E. K., & Al-Jarallah, R. A. (2005). Marshall--Olkin extended Weibull distribution and its application to censored data. Journal of Applied Statistics, 32, 1025–1034.
Ghitany, M. E., Atieh, B., & Nadarajah, S. (2008). Lindley distribution and its application. Mathematics and Computers in Simulation, 78, 493–506.
Gui, W. (2013). Marshall-Olkin extended log-logistic distribution and its application in minification processes. Applied Mathematical Sciences, 7, 3947–3961.
Inlow, M. (2010). A moment generating function proof of the Lindebergy central limit theorem. The American Statistician, 64, 228–230.
Jeong, B. Y., Murshed, M. S., Am Seo, Y., & Park, J. S. (2014). A three-parameter kappa distribution with hydrologic application: A generalized gumbel distribution. Stochastic Environmental Research and Risk Assessment, 28, 2063–2074.
Jose, K. K. (2011). Marshall-Olkin family of distributions and their applications in reliability theory, time series
Page 19 of 19
Okorie et al., Cogent Mathematics (2017), 4: 1285093http://dx.doi.org/10.1080/23311835.2017.1285093
© 2017 The Author(s). This open access article is distributed under a Creative Commons Attribution (CC-BY) 4.0 license.You are free to: Share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the material for any purpose, even commercially.The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. No additional restrictions You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Cogent Mathematics (ISSN: 2331-1835) is published by Cogent OA, part of Taylor & Francis Group. Publishing with Cogent OA ensures:• Immediate, universal access to your article on publication• High visibility and discoverability via the Cogent OA website as well as Taylor & Francis Online• Download and citation statistics for your article• Rapid online publication• Input from, and dialog with, expert editors and editorial boards• Retention of full copyright of your article• Guaranteed legacy preservation of your article• Discounts and waivers for authors in developing regionsSubmit your manuscript to a Cogent OA journal at www.CogentOA.com
modeling and stress-strength analysis. Proceeding 58th World Statistics Congress of the International Statistical Institute, Dublin (Session CPS005), 21st-26th August (pp. 3918–3923).
Jose, K. K., & Krishna, E. (2011, October). Marshall-Olkin extended uniform distribution. ProbStat Forum, 4, 78–88.
Mansoor, M., Tahir, M. H., Alzaatreh, A., Cordeiro, G. M., Zubair, M., & Ghazali, S. S. (2016). An extended Fréchet distribution: Properties and applications. Journal of Data Science, 14, 167–188.
Marshall, A. W., & Olkin, I. (1997). A new method for adding a parameter to a family of distributions with application to the exponential and Weibull families. Biometrika, 84, 641–652.
McLeish, D. (2014). Simulating random variables using moment-generating functions and the saddlepoint approximation. Journal of Statistical Computation and Simulation, 84, 324–334.
Meintanis, S. G. (2010). Testing skew normality via the moment generating function. Mathematical Methods of Statistics, 19, 64–72.
Merovci, F., & Elbatal, I. (2013). Transmuted Lindley-geometric distribution and its applications. arXiv preprint arXiv:1309.3774.
Moors, J. J. A. (1986). The meaning of kurtosis: Darlington reexamined. The American Statistician, 40, 283–284.
Nadarajah, S. (2008). Marshall and Olkins distributions. Acta Applicandae Mathematicae, 103, 87–100.
Oluyede, B., Foya, S., Warahena-Liyanage, G., & Huang, S. (2016). The log-logistic Weibull distribution with applications to lifetime data. Austrian Journal of Statistics, 45, 43–69.
Pogány, T. K., Saboor, A., & Provost, S. (2015). The Marshall-Olkin exponential Weibull distribution. Hacettepe Journal of Mathematics and Statistics.
RistiĆ, M. M., Jose, K. K., & Ancy, J. (2007). A Marshall--Olkin gamma distribution and minification process. Stress and Anxiety Research Society, 11, 107–117.
RistiĆ, M. M., & Kundu, D. (2015). Marshall-Olkin generalized exponential distribution. METRON, 73, 317–333.
Sandhya, E., & Prasanth, C. B. (2014). Marshall-Olkin discrete uniform distribution. Journal of Probability, 2014, 1–10.
Song, K. S. (2001). Rènyi information, loglikelihood and an intrinsic distribution measure. Journal Of Statistical Planning And Inference, 93, 51–69.
Sugiura, N. (1978). Further analysts of the data by akaike’s information criterion and the finite corrections: Further analysts of the data by akaike’s. Communications in Statistics-Theory and Methods, 7, 13–26.
Villa, E. R., & Escobar, L. A. (2006). Using moment generating functions to derive mixture distributions. The American Statistician, 60, 75–80. doi:10.1198/000313006X90819
von Waldenfels, W. (1987). Proof of an algebraic central limit theorem by moment generating functions (pp. 342–347). Berlin Heidelberg: Springer.