mathematics

Article
Study of a Modified Kumaraswamy Distribution
Rashad A. R. Bantan 1 , Christophe Chesneau 2, *, Farrukh Jamal 3 , Mohammed Elgarhy 4 , Waleed Almutiry 5
and Amani Abdullah Alahmadi 6

1 Department of Marine Geology, Faculty of Marine Science, King Abdulaziz University,

Jeddah 21551, Saudi Arabia; [email protected]
2 Department of Mathematics, Campus II, Université de Caen Normandie, Science 3, 14032 Caen, France
3 Department of Statistics, The Islamia University of Bahawalpur, Bahawalpur 63100, Pakistan;
[email protected]
4 The Higher Institute of Commercial Sciences, Al Mahalla Al Kubra, Al Gharbia 31951, Egypt;
[email protected]
5 Department of Mathematics, College of Science and Arts in Ar Rass, Qassim University, Buryadah 52571,
Saudi Arabia; [email protected]
6 College of Science and Humanities, Shaqra University, Shaqra 15572, Saudi Arabia; [email protected]
* Correspondence: [email protected]

Abstract: In this article, a structural modification of the Kumaraswamy distribution yields a new
two-parameter distribution defined on (0, 1), called the modified Kumaraswamy distribution. It has
the advantages of being (i) original in its definition, mixing logarithmic, power and ratio functions,
(ii) flexible from the modeling viewpoint, with rare functional capabilities for a bounded distribution—



in particular, N-shapes are observed for both the probability density and hazard rate functions—


and (iii) a solid alternative to its parental Kumaraswamy distribution in the first-order stochastic sense.
Citation: Bantan, R.A.R.; Chesneau, Some statistical features, such as the moments and quantile function, are represented in closed form.
C.; Jamal, F.; Elgarhy, M.; Almutiry, The Lambert function and incomplete beta function are involved in this regard. The distributions of
W.; Alahmadi, A.A. Study of a order statistics are also explored. Then, emphasis is put on the practice of the modified Kumaraswamy
Modified Kumaraswamy model in the context of data fitting. The well-known maximum likelihood approach is used to
Distribution. Mathematics 2021, 9,
estimate the parameters, and a simulation study is conducted to examine the performance of this
2836. https://doi.org/10.3390/
approach. In order to demonstrate the applicability of the suggested model, two real data sets are
math9212836
considered. As a notable result, for the considered data sets, statistical benchmarks indicate that
the new modeling strategy outperforms the Kumaraswamy model. The transmuted Kumaraswamy,
Academic Editors: Francisco De Asís
Torres-Ruiz and Antonio Barrera
beta, unit Rayleigh, Topp–Leone and power models are also outperformed.
García
Keywords: Kumaraswamy distribution; logarithmic transformation; moments; quantile; real data
Received: 25 September 2021 applications
Accepted: 5 November 2021
Published: 8 November 2021 MSC: 60E05; 62E15; 62F10

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in
published maps and institutional affil- 1. Introduction
iations. The perfect modeling of random processes that one regularly observes in applied
sciences remains in the order of utopia. Scientists and practitioners have already discussed
a wide range of models for solving these challenges. When the processes take values in
the interval (0, 1), a primary statistical analysis can be performed with the standard beta
Copyright: © 2021 by the authors. and Kumaraswamy (Kw) models. Before going further, a retrospective on these models
Licensee MDPI, Basel, Switzerland. is necessary. First, they are derived from the beta and Kw distributions, respectively.
This article is an open access article The beta distribution is a continuous distribution with support on the interval (0, 1) and
distributed under the terms and parameterized by two positive shape parameters, α and β. It is mathematically defined by
conditions of the Creative Commons the following cumulative distribution function (cdf):
Attribution (CC BY) license (https://
Z x
creativecommons.org/licenses/by/ 1
Fbeta ( x; α, β) = tα−1 (1 − t) β−1 dt, x ∈ (0, 1),
4.0/). B(α, β) 0

Mathematics 2021, 9, 2836. https://doi.org/10.3390/math9212836 https://www.mdpi.com/journal/mathematics

Mathematics 2021, 9, 2836 2 of 26

R1
where B(α, β) = 0 tα−1 (1 − t) β−1 dt, Fbeta ( x; α, β) = 0 for x ≤ 0, and Fbeta ( x; α, β) = 1 for
x ≥ 1. Naturally, the values of α and β influence the shape of the distribution. The cor-
responding probability density function (pdf) can be U-shaped, bell-shaped, monotonic
(increasing or decreasing), or even straight lines. On the other hand, the corresponding
hazard rate function (hrf) can be increasing with convex shapes, or U-shaped (see [1,2]). In a
wide number of areas, the beta distribution is used to model the behavior of characteristics
representing percentages and proportions. See, for example, the developments in [3].
In [4], the Kw distribution was introduced to complement the beta distribution. Thus,
it is also a continuous distribution with support in (0, 1) and parameterized by two positive
shape parameters, α and β. The Kw distribution is conceptually based on uniform order
statistics, and it has extremely simple functions that do not rely on special functions. The cdf
of the Kw distribution is given as follows:

FKw ( x; α, β) = 1 − (1 − x β )α , x ∈ (0, 1),

where FKw ( x; α, β) = 0 for x ≤ 0, and FKw ( x; α, β) = 1 for x ≥ 1. The form of the

functions related to this distribution is naturally influenced by the values of α and β. Its
application domain is identical to that of the beta distribution. As a brief comparison,
the Kw distribution is similar to the beta distribution, but it is much easier to work with
thanks to the tractability of its functions. We refer the reader to [5] and the references
therein for more information on this claim. As a fact, the beta and Kw models rarely
achieve the goal of optimal performance in all cases. For these reasons, new models
on the unit intervals have been proposed, derived from new distributions with support
equal to (0, 1). These distributions are more or less sophisticated in the analytical sense.
Contemporary works on this topic can be found in [6–19].
In this article, we propose a new two-parameter unit distribution, called the modified
Kumaraswamy (MKw) distribution. The considered cdf is such that, for x ∈ (0, 1), the
following holds: h i
FMKw ( x; α, β) = 1 − T (1 − x β )α ,

where T ( x ) denotes a certain analytical transformation that will be presented later. At this
introductory step, some features of T ( x ) are listed below.

Feature I. First of all, the transformation T ( x ) has an original form involving polynomial
and logarithmic functions without additional parameters. Thanks to it, some func-
tional capabilities of the Kw distribution are extended or modified, and can be more
adapted in some practical situations. In particular, the pdf of the MKw distribution
can be monotonic in a nearly angular way, “strongly” left- or right-skewed and of a
leptokurtic nature, or N-shaped. To our knowledge, these properties are not observed
for the pdf of the Kw distribution. On the other hand, the corresponding hrf presents
a wide panel of non-monotonic shapes, including N-shapes, that the hrf of the Kw
distribution does not possess. All these curvature differences give credit for a more
in-depth investigation of the MKw distribution from the modeling viewpoint.
Feature II. The transformation T ( x ) increases with respect to x. Thus, the intrinsic first-
order stochastic dominance (FOS) properties of the Kw distribution can be trans-
posed to the MKw distribution: for any 0 < α1 ≤ α2 and 0 < β 1 ≤ β 2 , we have
FMKw ( x; α1 , β 2 ) ≤ FMKw ( x; α2 , β 1 ) for any x ∈ R.
Feature III. The transformation T ( x ) satisfies the following inequality: T ( x ) ≥ x for any
x ∈ (0, 1). Thanks to it, the Kw and MKw distributions are involved in a simple
FOS dominance result: FMKw ( x; α, β) ≤ FKw ( x; α, β) for any x ∈ R. This means that
the MKw distribution generates a two-parameter statistical model, which does not
intersect with the Kw model in the cdf sense.

In fact, several mathematical interpretations of the MKw distribution are possible.

For instance, the MKw distribution may be viewed as a special parametric weighted
Mathematics 2021, 9, 2836 3 of 26

version of the Kw distribution. Additionally, it is deeply connected with the ratio power-
logarithmic (RPL) distribution created by [20]. Thanks to this connection, some proper-
ties of the RPL distribution may be used to determine those of the MKw distribution,
among other things. These diverse interpretations are detailed in the first part of the paper,
along with some theoretical properties. These properties include detailed quantile and
moment analyses, illustrated numerically and graphically for some selected values of α
and β. The order statistics are also discussed. The second part is devoted to the modeling
characteristics of the MKw distribution. The MKw model is considered, assuming that the
parameters α and β are unknown, and may be estimated via available data. We examine the
estimation of these parameters by the maximum likelihood method. We show the efficiency
of the obtained estimates by the use of simulated data. Then, two different data sets are
analyzed, and comparisons are made between the fits of the NKw, Kw, and beta models
and four other referenced models of the literature. We show that the proposed MKw
model outperforms the concurrence through the use of standard statistical benchmarks.
The findings are illustrated by numerous graphics.
The organization of the paper is composed of the following section. Section 2 intro-
duces the MKw distribution with full details, along with a functional analysis of the pdf
and hrf. Section 3 is devoted to quantile analysis and moment analysis. The statistical
inference on the parameters is considered in Section 4. Section 5 is devoted to applications.
We end the paper with a concluding part in Section 6.

2. The Modified Kw Distribution

This section focuses on the MKw distribution and its basic functional features.

2.1. Primary Definition

The following mathematical result is at the basis of the MKw distribution.

Proposition 1. Let α > 0 and β > 0, and G ( x; α, β) be the function defined on R by

(1 − x β ) α − 1
G ( x; α, β) = 1 − , x ∈ (0, 1),
α ln(1 − x β )

G ( x; α, β) = 0 for x ≤ 0, and G ( x; α, β) = 1 for x ≥ 1. Thus defined, G ( x; α, β) has the required

properties of a continuous cdf.

Proof. First, it is clear that G ( x; α, β) is continuous on R/{0, 1}. Let us examine its behavior
at the particular points x = 0 and x = 1. At the neighborhood of x = 0, by virtue of
standard equivalence functions, we have the following:

(1 − x β ) α − 1 −αx β
G ( x; α, β) = 1 − ∼ 1 − = 0 = G (0; α, β),
α ln(1 − x β ) α(− x β )

and, at the neighborhood of x = 1, we have

(1 − x β ) α − 1 1
G ( x; α, β) = 1 − ∼ 1+ → 1 = G (1; α, β).
α ln(1 − x )β α ln(1 − x β )

Therefore, G ( x; α, β) is continuous on {0, 1}, and on R. By construction, we have

limx→−∞ G ( x; α, β) = 0 and limx→+∞ G ( x; α, β) = 1. Let us now prove that G ( x; α, β)
is increasing. By applying standard rules of differentiation, for x ∈ (0, 1), we obtain
the following:

βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
 
0
G ( x; α, β) = . (1)
α( x β − 1)[ln(1 − x β )]2
Mathematics 2021, 9, 2836 4 of 26

The famous logarithmic inequality ln(1 + y) ≥ y/(1 + y) for y > −1, with y =
(1 − x β )α − 1 > −1 yields the following:
h i (1 − x β ) α − 1
α ln(1 − x β ) = ln 1 + (1 − x β )α − 1 ≥ = 1 − (1 − x β ) − α . (2)
1 + (1 − x β ) α − 1

This inequality can be rearranged as (1 − x β )α (1 − α ln(1 − x β )) − 1 ≤ 0. Therefore,

the numerator in Equation (1) is negative. Since the denominator is also negative, we
have G 0 ( x; α, β) ≥ 0, implying that G ( x; α, β) is increasing. This concludes the proof of
Proposition 1.

To the best of our knowledge, the cdf presented in Proposition 1 is new in the literature.
It defines a modified version of the Kw distribution, described as the MKw distribution,
discussed in Section 1. Some mathematical interpretations of this distribution are important
to understand why the term “modified” is employed here. This is developed in the
next section.

2.2. Mathematical Interpretations

Thus, the cdf of the MKw distribution is defined by the following:

(1 − x β ) α − 1
FMKw ( x; α, β) = 1 − , x ∈ (0, 1), (3)
α ln(1 − x β )

where FMKw ( x; α, β) = 0 for x ≤ 0, and FMKw ( x; α, β) = 1 for x ≥ 1, with α > 0 and β > 0.
Some mathematical interpretations of this cdf are given below.

Interpretation 1 As sketched in Section 1, for x ∈ (0, 1), we can write the following:
h i
FMKw ( x; α, β) = 1 − T (1 − x β )α ,

with T ( x ) = ( x − 1)/ ln( x ). In this sense, the transformation T ( x ) modifies the

mathematical structure of the cdf of the Kw distribution for different modeling per-
spectives.
The following facts can be deduced from the well-known logarithmic inequality
ln(1 + y) ≥ y/(1 + y) for y > −1 when applied to y = x − 1:
• T 0 ( x ) = ( x ln( x ) + 1 − x )/[ x (ln( x ))2 ] ≥ 0. This implies that T ( x ) is an increas-
ing function with respect to x. Following this is the stochastic FOS property: for
any 0 < α1 ≤ α2 and 0 < β 1 ≤ β 2 , we have FMKw ( x; α1 , β 2 ) ≤ FMKw ( x; α2 , β 1 )
for any x ∈ R.
• T ( x ) = ( x − 1)/ ln( x ) ≥ x for any x ∈ (0, 1). This implies that FMKw ( x; α, β) ≤
FKw ( x; α, β) for any x ∈ R. In other words, there is a direct FOS dominance of
the MKw distribution over the Kw distribution.
At the neighborhood of x = 0, we have T ( x ) ∼ −1/ ln( x ) → 0. This convergence is
slow, compared to the polynomial convergence to 0. This asymptotic result plays an
important role to understand the possible difference in the right skewness between
the MKw and Kw distributions; without more investigation, from the statistical
viewpoint, it is natural to think that the MKw model is more able to capture the slow
right-decay of a phenomena that the Kw model. This will be confirmed later with a
skewness analysis through quantile and moment methods, as well as practical data.
Interpretation 2 As a twin viewpoint, for x ∈ (0, 1), we can write the following:

FMKw ( x; α, β) = 1 − w( x; α, β)(1 − x β )α ,
Mathematics 2021, 9, 2836 5 of 26

with w( x; α, β) = [1 − (1 − x β )−α ]/[α ln(1 − x β )]. Thus, the weight function w( x; α, β)

modulate the main polynomial term of the cdf of the Kw distribution. In this sense,
the MKw distribution is a weighted version of the Kw distribution.
Interpretation 3 There is a deep relationship between the MKw and RPL distributions.
First, we recall that the RPL distribution is defined by the following logarithmic-
power cdf:

xα − 1
FRPL ( x; α) = , x ∈ (0, 1),
α ln( x )

FRPL ( x; α) = 0 for x ≤ 0 and FRPL ( x; α) = 1 for x ≥ 1. Then, for a random variable

Y following the RPL distribution, the random variable X = (1 − Y )1/β follows the
MKw distribution. Indeed, for x ∈ (0, 1), we have the following:

P( X ≤ x ) = P((1 − Y )1/β ≤ x ) = P(1 − Y ≤ x β ) = P(Y ≥ 1 − x β )

= 1 − FRPL (1 − x β ; α) = FMKw ( x; α, β).

By identification, the stated result is obtained. Thanks to the representation of X,

some properties of the RPL distribution can be used to determine those of the MKw
distribution.

Finally, because any random variable X defined on (0, 1) can be shifted to a bounded
support of the form ( a, b) with a < b, using the transformation (b − a) X + a, the MKw
distribution can easily be extended to any bounded domain.

2.3. Functional Analysis

This part is devoted to a functional analysis of the main functions of the MKw distri-
bution. To begin, we recall that the cdf of the MKw distribution is given by Equation (3).
As any cdf, it is an increasing function. Some asymptotic results are now presented. In the
neighborhood of x = 0, by virtue of standard equivalence functions at orders of 1 or 2,
we have the following:

(1 − x β ) α − 1 −αx β + α(α − 1) x2β /2 α

FMKw ( x; α, β) = 1 − ∼ 1− 2
= xβ.
α ln(1 − x )β α(− x ) − α(− x ) /2
β β 2

Thus, FMKw ( x; α, β) converges to 0 with a polynomial decay. In the neighborhood of

x = 1, we have the following:

(1 − x β ) α − 1 1
FMKw ( x; α, β) = 1 − ∼ 1+ .
α ln(1 − x )β α ln(1 − x β )

This implies that FMKw ( x; α, β) converges to 1 with a slow rate of convergence, due to
the logarithmic term.
Let us now focus our attention on the pdf and hrf of the MKw distribution. First,
the pdf is given by the following:

βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
 
0
f MKw ( x; α, β) = FMKw ( x; α, β) = , x ∈ (0, 1),
α( x β − 1)[ln(1 − x β )]2

and f MKw ( x; α, β) = 0 for x 6∈ (0, 1). Because of its analytical complexity, this function is
difficult to investigate using ordinary mathematical techniques (derivatives, . . . ). We thus
propose to study its asymptotic behaviors, then provide a graphical analysis to reveal its
Mathematics 2021, 9, 2836 6 of 26

possible shapes. At the neighborhood of x = 0, with similar arguments as those used for
the cdf, we have the following:

βα β−1
f MKw ( x; α, β) ∼ x .
2
From this result, it comes that f MKw ( x; α, β) explodes to +∞ for β < 1, is equal to the
constant α/2 for β = 1, and tends to 0 for β > 1. The values of β are thus of particular
importance for the limit points. At the neighborhood of x = 1, we have the following:

β 1
f MKw ( x; α, β) ∼ .
α (1 − x )[ln(1 − x β )]2
β

Thus, f MKw ( x; α, β) explodes to +∞ for all the values of α > 0 and β > 0, more or
less slowly. We complete this first approach by a graphical analysis. Figure 1 reveals a
maximum of different shapes that possesses f MKw ( x; α, β).
4

4
α = 10 β = 0.5 α = 0.2 β = 5
α = 10 β = 1.5 α = 0.8 β = 0.3
α = 10 β = 2.8 α = 6 β = 0.7
α = 10 β=5 α=5 β=6
3

3
pdf

pdf
2

2
1

1
0

0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8 1.0

x x

(a) (b)
Figure 1. Plots showing different shapes of the pdf of the MKw distribution: (a) decreasing and
unimodal shapes and (b) U- and N-shapes.

From Figures 1, we remark that f MKw ( x; α, β) can be decreasing, unimodal with

all the possible skewed directions, and also can be U- and N-shapes, which remains a
rare property for a distribution supported on (0, 1). In particular, the almost angular
monotonicity, “strongly” left-or right-skewed and N shape properties are not immediate
for the Kw distribution. To this first graphical analysis, we propose to show the versatility
of f MKw ( x; α, β) via three-dimensional (3D) plots in Figures 2 and 3, with moving α for
Figure 2, and moving β for Figure 3.

2.0

1.5
1.5
pdf

pdf

1.0
1.0

0.5 2.0 0.5 2.0

1.5 1.5
ha

ha

0.0 0.0
alp

alp

0.2 1.0 0.2 1.0

0.4 0.4
x 0.6 0.5 x 0.6 0.5
0.8 0.8
1.0 1.0

(a) (b)
Figure 2. Cont.
Mathematics 2021, 9, 2836 7 of 26

2.5
2.0

2.0
1.5

pdf
pdf
1.5
1.0
1.0

0.5 2.0 2.0

0.5
1.5 1.5

ha

ha
0.0 0.0

alp

alp
0.2 1.0 0.2 1.0
0.4 0.4
x 0.6 0.5 x 0.6 0.5
0.8 0.8
1.0 1.0

(c) (d)
Figure 2. The 3D plots of the pdf of the MKw distribution for α ∈ [1, 2] with (a) β = 0.5, (b) β = 0.9,
(c) β = 1.5, and (d) β = 2.5.

2.0
2.0

1.5
1.5
pdf

pdf
1.0
1.0

0.5 2.0 2.0

0.5
1.5 1.5
ta

ta
0.0 0.0
be

be
0.2 1.0 0.2 1.0
0.4 0.4
x 0.6 0.5 x 0.6 0.5
0.8 0.8
1.0 1.0

(a) (b)

2.0

2.0

1.5
1.5
pdf

pdf

1.0
1.0

0.5 2.0 2.0

0.5
1.5 1.5
ta

ta
0.0 0.0
be

be
0.2 1.0 0.2 1.0
0.4 0.4
x 0.6 0.5 x 0.6 0.5
0.8 0.8
1.0 1.0

(c) (d)
Figure 3. The 3D plots of the pdf of the MKw distribution for β ∈ [1, 2] with (a) α = 0.5, (b) α = 0.9,
(c) α = 1.5, and (d) α = 2.5.

The hrf of the MKw distribution is given by the following:

f MKw ( x; α, β)
h MKw ( x; α, β) =
1 − FMKw ( x; α, β)
βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
 
= , x ∈ (0, 1),
( x β − 1) ln(1 − x β )[(1 − x β )α − 1]

and h MKw ( x; α, β) = 0 for x 6∈ (0, 1). This function is difficult to explore using traditional
mathematical tools, due to its analytical complexity. We proceed in the same manner as for
f MKw ( x; α, β). In the neighborhood of x = 0, we have the following:

βα β−1
h MKw ( x; α, β) ∼ x .
2
As a result, h MKw ( x; α, β) explodes to +∞ for β < 1 (and thus cannot be increasing),
is equal to the constant α/2 for β = 1, and tends to 0 for β > 1. The values of β are thus
Mathematics 2021, 9, 2836 8 of 26

of particular importance for the limit points. In the neighborhood of x = 1, we obtain

the following:
1
h MKw ( x; α, β) ∼ − β .
(1 − x β ) ln(1 − x β )
Hence, h MKw ( x; α, β) exploses to +∞ for all the values of α > 0 and β > 0. A graphical
analysis now completes these results, with a focus on the possible shapes of h MKw ( x; α, β).
The maximum number of distinct shapes of h MKw ( x; α, β) can be seen in Figure 4.

α=2 β=8 α = 5 β = 0.8

4

4
α = 0.8 β = 1.2 0.4 β = 0.4
α=5 β=2 α = 15 β = 0.001
3

3
hrf

hrf
2

2
1

1
0

0
0.0 0.2 0.4 0.6 0.8 0.0 0.2 0.4 0.6 0.8

x x

(a) (b)
Figure 4. Plots showing different shapes of the hrf of the MKw distribution: (a) increasing, convex,
constant and N-shapes, and (b) U-shapes.

From Figure 4, it is clear that h MKw ( x; α, β) can be increasing with various convex
properties, and can be non-monotonic with N- or U- shapes. Again, these remain as rare
features for a distribution supported on (0, 1). For instance, the hrf of the Kw distribution
does not possess an N shape.
To illustrate the adaptability of h MKw ( x; α, β), we propose to use 3D plots in Figures 5 and 6,
with moving α for Figure 5 and moving β for Figure 6.

4
4

3
3
hrf
hrf

2
2
1.0 1.0
1 1
0.8 0.8

0.0 0.6 0.0 0.6

ha

ha
alp

alp

0.2 0.2
0.4 0.4 0.4 0.4
x 0.6 x 0.6
0.8 0.2 0.8 0.2

1.0 1.0

(a) (b)

5
5

4
4
hrf

3
hrf

2 2
1.0 1.0
1 1
0.8 0.8

0.0 0.6 0.0 0.6

ha

ha
alp

alp

0.2 0.2
0.4 0.4 0.4 0.4
x 0.6 x 0.6
0.8 0.2 0.8 0.2

1.0 1.0

(c) (d)
Figure 5. The 3D plots of the hrf of the MKw distribution for α ∈ [1, 2] with (a) β = 0.5, (b) β = 0.9,
(c) β = 1.5, and (d) β = 2.5.
Mathematics 2021, 9, 2836 9 of 26

3
4

hrf
2

hrf
2
1 1.0 1.0
0.8 1 0.8

0.0 0.6 0.0 0.6

ta

ta
be

be
0.2 0.2
0.4 0.4 0.4 0.4
x 0.6 x 0.6
0.8 0.2 0.8 0.2

1.0 1.0

(a) (b)

6
6

5
5

4
4

hrf
hrf

3
3
2 1.0 1.0
2
0.8 0.8
1
0.0 0.6 0.0 0.6

ta

ta
be

be
0.2 0.2
0.4 0.4 0.4 0.4
x 0.6 x 0.6
0.8 0.2 0.8 0.2

1.0 1.0

(c) (d)
Figure 6. The 3D plots of the hrf of the MKw distribution for β ∈ [1, 2] with (a) α = 0.5, (b) α = 0.9,
(c) α = 1.5, and (d) α = 2.5.

3. Mathematical Analysis
The MKw distribution is now mathematically treated, with emphasis on the quantile
and moment features.

3.1. Quantile Analysis

The fact that the cdf of the KMw distribution is invertible is one of its most notable
features. The quantile function (qf) can be expressed in closed form in terms of the principal
branch of the Lambert function, as described in the next proposition. We may refer to [21]
for further details on the Lambert function.

Proposition 2. The qf of the MKw distribution is defined by the following:

   1/β
1
Q MKw (u; α, β) = α(1 − u)W0 − e−1/[α(1−u)] +1 , u ∈ (0, 1),
α (1 − u )

where W0 ( x ) denotes the principal branch of the Lambert function.

−1
Proof. The qf is defined by Q MKw (u; α, β) = FMKw (u; α, β). Therefore, we can determine
it by solving the following nonlinear equation: FMKw (y; α, β) = u with respect to y. We
proceed as follows:
Mathematics 2021, 9, 2836 10 of 26

(1 − y β ) α − 1 (1 − y β ) α − 1
u = FMKw (y; α, β) ⇔ u = 1− ⇔ = 1−u
α ln(1 − y β ) α ln(1 − y β )
(1 − y β ) α − 1
1 − y β = e (1− y
β )α / [ α (1− u )]
⇔ ln(1 − y β ) = ⇔ e−1/[α(1−u)]
α (1 − u )
1 − y β −(1−y β )α /[α(1−u)] 1
⇔ − e =− e−1/[α(1−u)]
α (1 − u ) α (1 − u )
1 − yβ
 
1 −1/[α(1−u)]
⇔ − = W0 − e
α (1 − u ) α (1 − u )
   1/β
1
⇔ y = α(1 − u)W0 − e−1/[α(1−u)] + 1 .
α (1 − u )

We obtain the expected result.

The expression of the qf is essential to determine the main quartle of the MKw
distribution, such as the median defined by the following:
     1/β
1 α 2
M = Q MKw ; α, β = W0 − e−2/α + 1 .
2 2 α

Similarly, the first and third quartiles of the MKw distribution are given by
Q1 = Q MKw (1/4; α, β) and Q3 = Q MKw (3/4; α, β). The octiles of the MKw distribu-
tion are given by O1 = Q MKw (1/8; α, β), O3 = Q MKw (3/8; α, β), O5 = Q MKw (5/8; α, β),
and O7 = Q MKw (7/8; α, β). These special quantile values can serve as the main tool to
describe the skewness and kurtosis of the distribution. For instance, we can measure the
skewness via the quantile coefficient of Bowley defined by the following:

M − Q1
QS = 1 − 2 .
Q3 − Q1

On the other hand, we can measure the kurtosis by the quantile coefficient of Moors
defined by the following:
O7 + O3 − (O1 + O5 )
QK = .
Q3 − Q1
The sign of QS indicates the skewness of the MKw distribution: if QS > 0, the distri-
bution is right-skewed; if QS < 0, the distribution is left-skewed; and if QS approaches 0,
the distribution is nearly symmetric. Concerning the coefficient of Moors, we can compare
it to the value of 1.233, which corresponds to the quantile coefficient of Moors associated
with the standard normal distribution. As a result, if QK < 1.233, the distribution is
platykurtic; if QK > 1.233, the distribution is leptokurtic; and if QK approaches 1.233,
the distribution is mesokurtic.
The 3D plots for QS and QK for varying α and β are displayed in Figure 7.
From Figure 7, we see that the behavior of the quantile coefficients of skewness and
kurtosis is consistent with what was determined in Figures 1–3: the MKw distribution
exhibits all sorts of skewness since QS might be positive, almost equal to 0, or positive.
Furthermore, the fact that QK can be less than, nearly equal to, or larger than 1.2333 demon-
strates that all kurtosis levels are attained.
Mathematics 2021, 9, 2836 11 of 26

Skewness Kurtosis

0.5 6
5
0.0 4
1.0 3 1.0
−0.5 2
0.8 0.8
1
0.6 0.6

ha

ha
1.0 1.0

alp

alp
0.8 0.8
0.4 0.4
be0.6
ta be0.6
ta
0.4 0.4
0.2 0.2
0.2 0.2

Figure 7. The 3D plots of QS and QK for (α, β) ∈ [0.1, 1]2 .

In addition, the expression of the qf is particularly useful for simulations. Indeed,

pseudo-random data from the MKw distribution may be easily computer-generated using
the inverse transform approach. Since the Lambert function is accessible in computer
systems such as Maple, MATLAB and R, this approach is quite feasible. Further details on
the above quantile analysis can be found in [22].

3.2. Moment Analysis

This section is devoted to the moment properties of the MKw distribution, begin-
ning with the raw moments. Hereafter, we designate by X a random variable with the
MKw distribution.

3.2.1. Raw Moments

For any integer r, the rth raw moment of X can be expressed in an integral form as
follows:
Z +∞ β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
Z 1  
r r r βx
m (r ) = E ( X ) = x f MKw ( x; α, β)dx = x dx.
−∞ 0 α( x β − 1)[ln(1 − x β )]2

This integral can be implemented in any mathematical software, and can be calculated
numerically for given values of α and β. Some mathematical developments of this integral
are proposed in the next proposition.

Proposition 3. Two representations of the rth moment of X are given below.

Representation 1 We can express m(r ) in terms of the beta function as the following:
Z α  
r r
m (r ) = B , t + 1 dt,
αβ 0 β
R1
where B( a, b) = 0 t a−1 (1 − t)b−1 dt, with a > 0 and b > 0.
Representation 2 We can express m(r ) as the following series expansion:
+∞    
r/β k  α
m (r ) = 1 + ∑ k
(−1)k 1 − ln 1 +
α k
.
k =1

Proof. Let us prove the two representations in turn.

Representation 1 First, by an integration by part, we have the following:

Z 1 Z 1
r −1 (1 − x β ) α − 1 r
m (r ) = r x [1 − FMKw ( x; α, β)]dx = r x r −1 dx = ψ(α, r ), (4)
0 0 α ln(1 − x β ) α
Mathematics 2021, 9, 2836 12 of 26

where
Z 1
(1 − x β ) α − 1
ψ(α, r ) = x r −1 dx,
0 ln(1 − x β )

with ψ(α, r ) = 0 for α = 0. Owing to the Leibnitz integral rule and the change of
variables y = x β , we have the following:

1
 Z 1
r −1 (1 −
x β )α − 1 r −1 ∂ (1 − x ) − 1
Z  β α 
∂ ∂
ψ(α, r ) = x dx = x dx
∂α ∂α 0 ln(1 − x β ) 0 ∂α ln(1 − x β )
Z 1
1 1 r/β−1
 
1 r
Z
r −1
= x (1 − x ) dx =
β α
y (1 − y) dy = B
α
,α+1 .
0 β 0 β β

Since ψ(α, r ) = 0 for α = 0, upon integration with respect to α, we obtain the

following:
 
1 α r
Z
ψ(α, r ) = B , t + 1 dt. (5)
β 0 β

The desired result follows by substituting Equation (5) in Equation (4).

Representation 2 We recall that, in the distribution sense, X can be written as X = (1 −
Y )1/β , where Y follows the RPL distribution. Hence, for any positive integer r,
the general binomial theorem yields the following infinite series representation:
+∞  
r/β
r
X = 1+ ∑ k
(−1)k Y k . (6)
k =1

This expansion is valid almost everywhere; the event {Y = 1} is of probability zero.

Now, let us present a known result (see [20] (Proposition 6)). We obviously have
E(Y 0 ) = 1, and, for any integer k ≥ 1, the kth raw moment of Y is obtained as
the following:

k  α
m ∗ ( k ) = E (Y k ) = 1 − ln 1 + .
α k

Therefore, based on Equation (6) and the linearity of the expectation operator, we have
the following:
+∞ 

r/β
m (r ) = 1 + ∑ k (−1)k m∗ (k)
k =1
+∞    
r/β k  α
= 1+ ∑ (−1)k 1 − ln 1 + .
k =1
k α k

The stated result is obtained.

The proof of Proposition 3 ends.

By arranging the raw moments in a suitable manner, we can determine the standard
central, dispersion, skewness, and kurtosis measures of X. These measures include the
mean given by m(1), the variance obtained through the Koenig–Huygens formula, Var =
E(( X − m(1))2 ) = m(2) − [m(1)]2 , the moment coefficient of skewness defined by

1 m(3) − 3m(1)Var − [m(1)]3

MS = 3/2
E(( X − m(1))3 ) = ,
Var Var3/2
Mathematics 2021, 9, 2836 13 of 26

and the moment coefficient of kurtosis given by the following:

1 m(4) − 4m(3)m(1) + 6m(2)[m(1)]2 − 3[m(1)]4

MK = 2
E(( X − m(1))4 ) = .
Var Var2
The interpretation of the moment skewness and kurtosis is similar to those of the
quantile moment and kurtosis. The only difference is that, for the kurtosis, the benchmark
value of the moment coefficient of kurtosis is 3 instead of 1.233 for the quantile coefficient
of kurtosis.
In addition, we can consider the coefficient of variation obtained as CV = Var1/2 /m(1)
and the index of dispersion given as ID = Var1/2 CV. Table 1 shows the numerical values
of m(1), m(2), m(3), m(4), Var, MS, MK, CV and ID for a random variable X following the
MKw distribution with α = 5 and varying values of β, while Table 2 does the same with
β = 3 and varying α.

Table 1. Numerical values of moment measures for a random variable X following the MKw
distribution with α = 5 and varying β.

β m (1) m (2) m (3) m (4) Var MS MK CV ID

0.5 0.894 0.835 0.793 0.76 0.036 −2.736 11.244 0.213 0.041
1.0 0.838 0.744 0.679 0.63 0.042 −1.796 6.393 0.245 0.05
1.5 0.798 0.676 0.602 0.545 0.039 −0.032 −6.622 0.248 0.049
2.0 0.762 0.631 0.543 0.477 0.05 −1.3 3.891 0.295 0.066
2.5 0.734 0.587 0.496 0.428 0.048 −0.534 −0.3 0.298 0.065
3.0 0.711 0.553 0.458 0.389 0.048 −0.419 0.127 0.309 0.068
3.5 0.69 0.525 0.424 0.357 0.049 −0.483 2.431 0.32 0.071
4.0 0.672 0.5 0.397 0.33 0.049 −0.38 2.341 0.329 0.073
4.5 0.655 0.478 0.374 0.308 0.049 −0.292 2.287 0.337 0.074

Table 2. Numerical values of moment measures for a random variable X following the MKw
distribution with β = 3 and varying α.

α m (1) m (2) m (3) m (4) Var MS MK CV ID

0.5 0.213 0.129 0.101 0.087 0.084 1.574 4.215 1.357 0.393
1.0 0.354 0.213 0.156 0.129 0.087 0.729 2.727 0.834 0.247
1.5 0.459 0.288 0.211 0.172 0.077 0.387 2.428 0.604 0.168
2.0 0.537 0.354 0.264 0.211 0.066 0.133 2.036 0.48 0.124
2.5 0.595 0.411 0.311 0.251 0.057 −0.084 2.195 0.401 0.096
3.0 0.641 0.459 0.354 0.288 0.049 −0.215 2.258 0.345 0.076
3.5 0.677 0.501 0.393 0.323 0.042 −0.377 2.749 0.303 0.062
4.0 0.707 0.537 0.428 0.354 0.037 −0.503 2.951 0.272 0.052
4.5 0.732 0.568 0.459 0.384 0.032 −0.625 3.349 0.246 0.044

From both Tables 1 and 2, we observe that the mean is monotonic with respect to α
and β, which is not the case for the variance. Furthermore, wide variations of values are
seen for the mean, whereas the variance remains relatively stable and small. The behavior
of the moment coefficients of skewness and kurtosis are consistent with what is concluded
in Figures 1–3, as well as with the use of quantile coefficients of skewness and kurtosis;
because MK can be positive, nearly equal to 0 or positive, the MKw distribution has all
types of skewness. Moreover, since MK can be lower, almost equal or greater than 3, it
indicates that all the kurtosis states are reached. In addition, we see a strong versatility in
the numerical values of CV and ID.
All these comments are illustrated graphically in Figure 8 via 3D plots with varying α
and β.
Mathematics 2021, 9, 2836 14 of 26

0.20
0.6

0.5 0.18

varia
mean

nc
0.4

e
0.16

2.0 3.0
0.3
0.14
1.5 2.5

ta
0.2 0.2 2.0

be

ta
1.0

be
0.4 0.4
alp alp 1.5
ha 0.6 0.5 ha 0.6
0.8 0.8
1.0

(a) (b)

1.0
2.5

0.5
skew

kurto
2.0
ness

sis
0.0
3.0 1.5 3.0
−0.5
2.5 2.5

0.2 2.0 0.2 2.0

ta

ta
be

be
0.4 0.4
alp 1.5 alp 1.5
ha 0.6 ha 0.6
0.8 0.8
1.0 1.0

(c) (d)

1.6
0.6
1.4
0.5
1.2
ID
cv

1.0 0.4

3.0 3.0
0.8 0.3
2.5 2.5
0.6

0.2 2.0 0.2 2.0

ta

ta
be

be
0.4 0.4
alp 1.5 alp 1.5
ha 0.6 ha 0.6
0.8 0.8
1.0 1.0

(e) (f)
Figure 8. The 3D plots of the (a) mean, (b) variance, (c) moment skewness, (d) moment kurtosis,
(e) coefficient of variation and (f) index of dispersion for the MKw distribution for α ∈ [0.1, 0.9] and
β ∈ [1, 3].
In the next part, we complete this moment analysis with the expression of the mean
log moments, and the incomplete moments of X.

3.2.2. Other Kinds of Moments

The mean logarithm of X can be expressed in an integral form as follows:
Z +∞
mln = E(ln( X )) = ln( x ) f MKw ( x; α, β)dx
−∞
βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
Z 1  
= ln( x ) dx.
0 α( x β − 1)[ln(1 − x β )]2

This kind of moment appears in some estimation methods, and entropy. It has the
advantage of having values in R, as opposed to m(1), which has values in (0, 1). This
can be of interest for the construction of regression models, among other things. The
integral defining mln may be determined numerically for given values of α and β and can
be implemented in any mathematical software. In the following result, the mean logarithm
of X is stated as a series expansion.
Mathematics 2021, 9, 2836 15 of 26

Proposition 4. The mean logarithm of X is the following:

+∞
(−1)k−1
 
1 k  α
mln =
β ∑ k 1 − ln 1 +
α k
.
k =1

Proof. For this proof, we adopt the setting used in the proof of Representation 2 of Propo-
sition 3. By using the relationship between the MKw and RPL distributions, and the
logarithmic series expansion, we obtain the following:

1 1 +∞ (−1)k−1
mln = E(ln(1 − Y )) = ∑ m∗ (k )
β β k =1 k
1 +∞ (−1)k−1
 
k  α
= ∑ 1 − ln 1 + .
β k =1 k α k

The desired expansion is proved.

Now, let y ∈ [0, 1] and Xy be the random variable equal to X if X ≤ y, and 0 otherwise.
Then, the rth incomplete moment of X truncated at y can be expressed as the following:
β −1
 
Z y Z y
r βx (1 − x β )α (1 − α ln(1 − x β )) − 1
m(r )[y] = E( Xyr ) = r
x f MKw ( x; α, β)dx = x dx.
−∞ 0 α( x β − 1)[ln(1 − x β )]2

The incomplete moments are involved in a lot of probabilistic measures and functions,
such as various mean deviations, residual life-type functions and curves. The integral
defining m(r )[y] may be determined numerically for given values of α and β and can be
implemented in any scientific software. The following proposition proposes a mathematical
development of this integral.

Proposition 5. Two representations of the rth incomplete moment of X truncated at y are given below.

Representation 1 We can express m(r )[y] in terms of the incomplete beta function as the following:

r (1 − y
−1 β )α  
r r
Z α
m(r )[y] = −y + B β , t + 1 dt,
α ln(1 − y β ) αβ 0 y β
Rc
where Bc ( a, b) = 0 t a−1 (1 − t)b−1 dt, with a > 0, b > 0 and c ∈ [0, 1].
Representation 2 We can express m(r )[y] as the following series expansion:
+∞ 
yα − 1
  
r/β k
m(r )[y] = 1 + ∑ k
(−1)k yk
ln(y)
− {Ei[(α + k) ln(y)] − Ei[k ln(y)]} ,
α
k =1
Rx
where Ei( x ) = −∞ et /tdt, x ∈ R∗ .

Proof. Let us prove the two representations in turn.

Representation 1 First, by an integration by part, we obtain the following:

Z y
m(r )[y] = −yr [1 − FMKw (y; α, β)] + r xr−1 [1 − FMKw ( x; α, β)]dx
0
Z y
(1 − y β ) α − 1 (1 − x β ) α − 1
= − yr +r x r −1 dx
α ln(1 − y β ) 0 α ln(1 − x β )
(1 − y β ) α
−1 r
= − yr + ψ(α, r )[y], (7)
α ln(1 − y β ) α
Mathematics 2021, 9, 2836 16 of 26

where
Z y
(1 − x β ) α − 1
ψ(α, r )[y] = x r −1 dx,
0 ln(1 − x β )

with ψ(α, r )[y] = 0 for α = 0. By applying the Leibnitz integral rule and the change
of variables z = x β , we obtain the following:

y
 Z y
r −1 (1 −
x β )α − 1 r −1 ∂ (1 − x ) − 1
Z  β α 
∂ ∂
ψ(α, r )[y] = x dx = x dx
∂α ∂α 0 ln(1 − x β ) 0 ∂α ln(1 − x β )
Z y Z β
1 y r/β−1
 
1 r
= xr−1 (1 − x β )α dx = z (1 − z)α dz = By β ,α+1 .
0 β 0 β β

Since ψ(α, r )[y] = 0 for α = 0, upon integration with respect to α, we obtain

the following:
 
1 α r
Z
ψ(α, r )[y] = B β , t + 1 dt. (8)
β 0 y β

Substituting Equation (8) in Equation (7) yields the desired result.

Representation 2 To begin, let us introduce a known result (see [20] (Proposition 9)).
The kth incomplete moment of a random variable Y with the RPL distribution at y
with y ∈ [0, 1] is given by the following:

yα − 1 k
m∗ (k)[y] = yk − {Ei[(α + k) ln(y)] − Ei[k ln(y)]}.
ln(y) α

Hence, by using Equation (6), we have the following:

+∞ 
r/β
m(r )[y] = 1 + ∑ k (−1)k m∗ (k)[y]
k =1
+∞ 
yα − 1
  
r/β k
= 1+ ∑ (−1)k yk − {Ei[(α + k) ln(y)] − Ei[k ln(y)]} .
k =1
k ln(y) α

The stated result is obtained.

The proof of Proposition 5 ends.

The remainder of the work is devoted to the creation of a new data analysis paradigm
based on the MKw model.

3.3. Order Statistics

In this section, we cover the fundamentals of the order statistics of the MKw dis-
tribution. Let n be a positive integer and X1 , · · · , Xn be independent and identically
distributed random variables following the MKw distribution. We consider the ran-
dom variables X1:n , . . . , Xn:n defined as the ordered versions of X1 , · · · , Xn , such that
P( X1:n ≤ . . . ≤ Xn:n ) = 1. Then, for any j = 1, 2, . . . , n, the pdf of X j:n is given as
the following:

n!
f j:n ( x; α, β) = f ( x; α, β)[ FMKw ( x; α, β)] j−1 [1 − FMKw ( x; α, β)]n− j .
( j − 1)!(n − j)! MKw
Mathematics 2021, 9, 2836 17 of 26

That is, in an expanded form, we have the following:

βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
 
n!
f j:n ( x; α, β) = ×
( j − 1) ! ( n − j ) ! α( x β − 1)[ln(1 − x β )]2
 j −1  n− j
(1 − x β ) α − 1 (1 − x β ) α − 1

1− , x ∈ (0, 1),
α ln(1 − x β ) α ln(1 − x β )

and f j:n ( x; α, β) = 0 for x 6∈ (0, 1). In particular, if we focus on the two extreme order
statistics, the pdf of X1:n is given by
 n −1
βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1 (1 − x β )α − 1
 
f 1:n ( x; α, β) = n , x ∈ (0, 1),
α( x β − 1)[ln(1 − x β )]2 α ln(1 − x β )

and f 1:n ( x; α, β) = 0 for x 6∈ (0, 1), and the pdf of Xn:n is given by the following:
 n −1
βx β−1 (1 − x β )α (1 − α ln(1 − x β )) − 1
 
(1 − x β ) α − 1
f n:n ( x; α, β) = n 1− , x ∈ (0, 1),
α( x β − 1)[ln(1 − x β )]2 α ln(1 − x β )

and f n:n ( x; α, β) = 0 for x 6∈ (0, 1). Moment analysis is possible using these formulas.
However, due to the functional complexity of the pdfs, numerical evaluations appear to
necessitate the employment of computer mathematical tools.

4. Statistical Inference
The parameters α and β of the MKw distribution are now considered to be unknown
and must be calculated using data. As a result, the MKw model is examined.

4.1. Estimation Method

The maximum likelihood (MLL) technique is well adapted for this purpose. The
specifics of this estimation strategy are outlined below.
Let x1 , x2 , . . . , xn symbolize the n values lying into (0, 1) that are predicted to be seen
from a random variable X with the MKw distribution and unknown parameters α and β.
Then, the log-likelihood function for α and β is formulated as the following:
n
L(α, β) = ∑ ln[ f MKw (xi ; α, β)]
i =1
n n n h i
= n ln β − n ln α + ( β − 1) ∑ ln( xi ) − ∑ ln(1 − xi ) − 2 ∑ ln −ln(1 − xi β )
β

i =1 i =1 i =1
n h  i
+ ∑ ln 1 − (1 − xi ) β α
1 − α ln(1 − xi β ) .
i =1

The MLL estimates (MLLEs) of α and β, denoted by α̂ and β̂, are defined by the following:

(α̂, β̂) = argmax(α,β)∈(0,+∞)2 L(α, β).

These MLLEs fulfill the partial derivative equations listed below: ∂L(α, β)/∂α = 0,
and ∂L(α, β)/∂β = 0, where
n zα {ln( z )(1 − α ln( z )) + ln( z )}
∂L(α, β) n
=− +∑ i
i i i
∂α α i =1 ziα (1 − α ln(zi )) − 1

and

∂L(α, β) n n n
t n
ti n zα−1 ti ln(zi )
= + ∑ ln( xi ) + ∑ i + 2 ∑ + α2 ∑ α i ,
∂β β i =1 z
i =1 i
z ln(zi )
i =1 i
z (1 − α ln(zi )) − 1
i =1 i
Mathematics 2021, 9, 2836 18 of 26

where zi = 1 − xi β and ti = xi β ln( xi ). Due to the intricacy of these partial derivatives,

it is difficult to derive closed-form equations for α̂ and β̂. Computational methods can
be used to obtain precise numerical solutions. With the MMLEs, we can estimate all the
functions of the MKw model by insertion. In particular, f MKw ( x; α, β) and FMKw ( x; α, β)
may be naturally estimated by f MKw ( x; α̂, β̂) and FMKw ( x; α̂, β̂), respectively.
A two-dimensional normal distribution with the mean (α, β) and matrix of covariance
could be used to identify the asymptotic distribution of the random version of (α̂, β̂). To be
more specific, the matrix of covariance is given by D = I (α, β)−1 |(α,β)=(α̂,β̂) , where the
following holds:

∂2 L(α, β) ∂2 L(α, β)
 
 ∂α2 ∂α∂β 
I (α, β) = − .
 ∂2 L(α, β) 
.
∂β2

The matrix D can be computed quantitatively using mathematical techniques. We can

establish estimated confidence intervals (CoIs) for α and β at a particular level, such as
100(1 − q)% with q ∈ (0, 1). The associated lower bounds (CoI-LBs) and upper bounds (CoI-
UBs) of such intervals remain traditional; if we keep our eyes on the unknown parameter
α, these interval bounds are provided by CoI-LB = α̂ − zq SErα and CoI-UB = α̂ + zq SErα ,
where zq is defined by P(| Z | ≥ zq ) = q, with Z a random variable having the standard
Gaussian distribution, and SErα is a term that relates to the standard error (SEr) of α̂,
determined by the square root of the first diagonal element of D. For more theoretical and
practical achievements of the MLLEs, see [23].

4.2. Simulation
Here, we perform simulation work to measure the behavior of the constructed MLLEs.
In this situation, we employ the R program, which was created by [24]. The following steps
are taken into consideration:
1. We apply the inverse transform approach to produce 10,000 random samples with
values x1 , x2 , . . . , xn . That is, for any i = 1, . . . , n, the value xi is computed as xi =
Q MKw (ui ; α, β), where ui is a value generated from the uniform distribution on (0, 1).
2. Six different sample sizes are taken into account: n = 30, 50, 100, 200, 500, and 1000.
3. Throughout the vector of parameters (α, β), we consider seven possible configura-
tions.
4. For the examined configurations, the average MLLEs, mean square errors (MSErs),
CoI-LBs, CoI-UBs, and length (CoI-LEN) of the CoIs are computed at two distinct
levels (90% and 95%).
5. Tables 3–9 show the outcomes that were achieved.

Table 3. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.5,
β = 0.5) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
1.6622 0.3629 0.9792 2.3451 1.3659 0.8485 2.4759 1.6274
30 0.5311 0.0223 0.3457 0.7165 0.3708 0.3102 0.7520 0.4418
1.6082 0.1195 1.1166 2.0998 0.9833 1.0224 2.1940 1.1716
50 0.5268 0.0111 0.3848 0.6688 0.2840 0.3576 0.6960 0.3384
1.5114 0.0382 1.1943 1.8284 0.6341 1.1336 1.8891 0.7555
100 0.4998 0.0037 0.4037 0.5958 0.1922 0.3853 0.6142 0.2290
Mathematics 2021, 9, 2836 19 of 26

Table 3. Cont.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
1.5141 0.0180 1.2896 1.7386 0.4490 1.2466 1.7816 0.5350
200 0.5015 0.0018 0.4333 0.5697 0.1364 0.4203 0.5828 0.1625
1.4992 0.0060 1.3593 1.6392 0.2799 1.3325 1.6660 0.3335
500 0.4991 0.0006 0.4561 0.5421 0.0860 0.4479 0.5503 0.1024
1.5151 0.0039 1.4146 1.6155 0.2009 1.3954 1.6348 0.2394
1000 0.5042 0.0004 0.4735 0.5348 0.0613 0.4677 0.5407 0.0731

Table 4. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.5,
β = 1.2) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.5485 0.0109 0.3935 0.7034 0.3099 0.3639 0.7331 0.3692
30 1.3745 0.1687 0.7487 2.0004 1.2517 0.6288 2.1202 1.4914
0.5303 0.0073 0.4167 0.6439 0.2273 0.3949 0.6657 0.2708
50 1.3645 0.1538 0.8800 1.8489 0.9690 0.7872 1.9417 1.1545
0.5125 0.0023 0.4354 0.5896 0.1541 0.4207 0.6043 0.1836
100 1.2505 0.0428 0.9322 1.5687 0.6365 0.8713 1.6297 0.7583
0.5038 0.0011 0.4508 0.5569 0.1062 0.4406 0.5671 0.1265
200 1.2282 0.0216 1.0060 1.4504 0.4444 0.9634 1.4929 0.5295
0.4973 0.0005 0.4644 0.5302 0.0658 0.4581 0.5365 0.0784
500 1.1947 0.0099 1.0575 1.3319 0.2744 1.0313 1.3582 0.3269
0.4999 0.0002 0.4765 0.5233 0.0469 0.4720 0.5278 0.0558
1000 1.2039 0.0027 1.1062 1.3015 0.1953 1.0875 1.3202 0.2326

Table 5. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 1.2,
β = 0.5) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
1.2612 0.0859 0.8003 1.7221 0.9218 0.7120 1.8104 1.0983
30 0.4939 0.0137 0.3132 0.6746 0.3613 0.2786 0.7092 0.4305
1.2505 0.0858 0.8961 1.6048 0.7087 0.8283 1.6727 0.8444
50 0.5147 0.0131 0.3683 0.6610 0.2927 0.3403 0.6891 0.3487
1.2547 0.0323 1.0054 1.5041 0.4987 0.9577 1.5518 0.5941
100 0.5182 0.0049 0.4143 0.6222 0.2079 0.3944 0.6421 0.2477
1.2206 0.0099 1.0516 1.3896 0.3380 1.0192 1.4219 0.4028
200 0.5035 0.0014 0.4318 0.5752 0.1433 0.4181 0.5889 0.1708
1.2091 0.0050 1.1037 1.3145 0.2107 1.0835 1.3346 0.2511
500 0.5035 0.0007 0.4581 0.5489 0.0908 0.4494 0.5576 0.1082
1.2081 0.0015 1.1338 1.2824 0.1486 1.1196 1.2967 0.1771
1000 0.5055 0.0004 0.4733 0.5377 0.0644 0.4671 0.5439 0.0768
Mathematics 2021, 9, 2836 20 of 26

Table 6. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.8,
β = 0.5) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.8409 0.0326 0.5709 1.1109 0.5400 0.5192 1.1626 0.6434
30 0.5234 0.0217 0.3126 0.7342 0.4216 0.2722 0.7746 0.5024
0.8634 0.0248 0.6475 1.0793 0.4318 0.6062 1.1206 0.5144
50 0.5526 0.0157 0.3816 0.7235 0.3419 0.3489 0.7562 0.4073
0.8432 0.0083 0.6963 0.9901 0.2938 0.6681 1.0182 0.3501
100 0.5268 0.0048 0.4109 0.6428 0.2319 0.3887 0.6650 0.2764
0.8118 0.0024 0.7046 0.8989 0.1943 0.6860 0.9175 0.2315
200 0.5074 0.0016 0.4273 0.5875 0.1601 0.4120 0.6028 0.1908
0.8038 0.0020 0.7559 0.8817 0.1258 0.7439 0.8938 0.1499
500 0.5098 0.0015 0.4593 0.5603 0.1010 0.4496 0.5699 0.1203
0.8023 0.0009 0.7589 0.8456 0.0867 0.7506 0.8539 0.1033
1000 0.5011 0.0004 0.4658 0.5363 0.0706 0.4590 0.5431 0.0841

Table 7. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.8,
β = 0.8) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.8875 0.0505 0.5967 1.1783 0.5815 0.5411 1.2340 0.6929
30 0.8732 0.0513 0.5263 1.2201 0.6939 0.4598 1.2866 0.8268
0.8615 0.0244 0.6491 1.0739 0.4248 0.6084 1.1146 0.5062
50 0.8514 0.0345 0.5889 1.1139 0.5250 0.5386 1.1642 0.6255
0.8282 0.0114 0.6837 0.9726 0.2888 0.6561 1.0002 0.3441
100 0.8408 0.0162 0.6545 1.0271 0.3726 0.6188 1.0628 0.4440
0.8035 0.0028 0.7063 0.9007 0.1943 0.6877 0.9193 0.2316
200 0.8069 0.0056 0.6798 0.9340 0.2541 0.6555 0.9583 0.3028
0.8040 0.0014 0.7426 0.8653 0.1227 0.7309 0.8771 0.1462
500 0.8033 0.0021 0.7233 0.8832 0.1599 0.7080 0.8985 0.1905
0.7977 0.0006 0.7547 0.8407 0.0860 0.7465 0.8489 0.1024
1000 0.8008 0.0011 0.7443 0.8573 0.1130 0.7335 0.8681 0.1346

Table 8. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.5,
β = 1.5) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.5295 0.0103 0.3812 0.6777 0.2965 0.3529 0.7061 0.3532
30 1.7019 0.3967 0.9201 2.4837 1.5636 0.7704 2.6334 1.8630
0.5269 0.0061 0.4128 0.6410 0.2282 0.3909 0.6628 0.2719
50 1.6688 0.1760 1.0730 2.2646 1.1916 0.9589 2.3786 1.4197
0.5192 0.0026 0.4411 0.5973 0.1562 0.4262 0.6123 0.1861
100 1.5894 0.0557 1.1863 1.9925 0.8063 1.1091 2.0697 0.9606
0.5065 0.0010 0.4531 0.5600 0.1068 0.4429 0.5702 0.1273
200 1.5332 0.0327 1.2565 1.8100 0.5535 1.2035 1.8630 0.6595
Mathematics 2021, 9, 2836 21 of 26

Table 8. Cont.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.5006 0.0004 0.4674 0.5339 0.0665 0.4610 0.5403 0.0792
500 1.5192 0.0115 1.3450 1.6934 0.3484 1.3116 1.7267 0.4151
0.4998 0.0002 0.4764 0.5233 0.0469 0.4719 0.5278 0.0559
1000 1.5007 0.0057 1.3790 1.6224 0.2434 1.3557 1.6457 0.2900

Table 9. Numerical values of the MLLEs, MSErs, CoI-LBs, CoI-UBs, and CoI-LENs for (α = 0.5,
β = 2.0) of the MKw model.

90% 95%
n MLLE MSEr
CoI-LB CoI-UB CoI-LEN CoI-LB CoI-UB CoI-LEN
0.5206 0.0063 0.3754 0.6658 0.2904 0.3476 0.6936 0.3460
30 2.1889 0.3108 1.1732 3.2046 2.0313 0.9788 3.3991 2.4203
0.5237 0.0043 0.4117 0.6357 0.2239 0.3903 0.6571 0.2668
50 2.1641 0.2081 1.3904 2.9377 1.5474 1.2422 3.0859 1.8437
0.5095 0.0021 0.4334 0.5857 0.1524 0.4188 0.6003 0.1816
100 2.0274 0.1227 1.5102 2.5447 1.0345 1.4111 2.6437 1.2326
0.4988 0.0009 0.4464 0.5512 0.1048 0.4364 0.5612 0.1248
200 2.0031 0.0449 1.6395 2.3667 0.7272 1.5699 2.4363 0.8664
0.5019 0.0003 0.4685 0.5353 0.0668 0.4621 0.5417 0.0796
500 1.9800 0.0214 1.7531 2.2070 0.4538 1.7097 2.2504 0.5407
0.5014 0.0002 0.4778 0.5250 0.0472 0.4733 0.5296 0.0563
1000 1.9914 0.0072 1.8299 2.1529 0.3230 1.7990 2.1838 0.3849

From the tables above, we see that the values of MLLEs are near to the values of the
parameters provided in the settings as n rises. Furthermore, when n rises, the MSEr and
CoI-LEN values fall, as one would anticipate.

5. Application
It is critical to demonstrate how the MKw model may be used in practice. This is the
aim of this section.

5.1. Method
In the coming applications, the fits of the MKw model are compared with those of
some competitive models, such as the transmuted Kumaraswamy (TKw), Kumaraswamy
(Kw), beta (B), unit Rayleigh (UR), Topp–Leone (Topp), and Power models. The TKw
model has three parameters, whereas the other models have two parameters only. These
models are defined through their cdfs in Table 10.

Table 10. Competent models with the MKw model.

Models Cdfs References

TKw [1 − (1 − x α )θ ][1 + λ(1 − x α )θ ] [6]
Kw 1Z− (1 − x λ )θ [4]
1 x
Beta tα1 −1 (1 − t)α2 −1 dt [1]
B ( α1 , α2 ) 0
2
UR e−τ [ln( x)] [12]
Topp [ x (2 − x )]µ [25]
Power xν (Classic)
Mathematics 2021, 9, 2836 22 of 26

In order to compare these models, we consider well-referenced statistical measures,

namely the Cramér–Von Mises (W), Anderson–Darling (A), Kolmogorov–Smirnov (KS),
and Akaike information criterion (AIC) based on the minus maximal likelihood value,
denoted by −`ˆ . The best model is the one with the smallest values of these statistics and
criteria. We refer to [26] for more information on the usage and underlying meaning of the
measures W, A, KS and AIC. In addition, we extract the p-value associated with the KS test.
The best model is the one with the largest p-value.
As for the simulated part, the results are obtained using the R software, and two real
data sets are considered.

5.2. Maximum Flood Level Data Set

The first data set comes from [27] and contains environmental data. It refers to 20
observations of the Susquehanna River’s greatest flood level (measured in millions of cubic
feet per second) in Harrisburg, Pennsylvania. The data are 0.26, 0.27, 0.30, 0.32, 0.32, 0.34,
0.38, 0.38, 0.39, 0.40, 0.41, 0.42, 0.42, 0.42, 0.45, 0.48, 0.49, 0.61, 0.65, 0.74. A histogram
analysis reveals that the data are unimodal and right skewed, with a small left-skewed tail.
Theoretically, in view of the shape ability of the related pdf, the MKw model is able to fit
these data.
Following our methodology, we first determine the MLLEs and the related SEs of the
considered models in Table 11.

Table 11. MLLEs and SEs of the models for the maximum flood level data set.

Models MLLEs (SEs)

MKw 354.0108 5.9002
(α, β) (5.0365) (1.2035)
TKw 3.7445 11.1649 0.6201
(α, θ, λ) (0.6505) (6.1634) (0.3722)
Kw 3.3773 12.0018
(λ, θ ) (0.6041) (5.4713)
Beta 6.8309 9.2364
( α1 , α2 ) (2.1179) (2.8912)
UR 1.1366
(τ ) (0.2541)
Topp 2.2412
(µ) (0.5011)
Power 1.7804
(ν) (0.2487)

From Table 11, if we focus on the MKw model, we see that α̂ = 354.0108 and β̂ = 5.9002.
By using these estimates, we derive function estimates for the pdf and cdf of the MKw
model through the substitution method; f MKw ( x; α, β) and FMKw ( x; α, β) may be naturally
estimated by f MKw ( x; α̂, β̂) and FMKw ( x; α̂, β̂), respectively.
The models are compared via our statistical benchmarks in Table 12.

Table 12. Values of −`ˆ , AIC, W, A, KS and (KS p-value) of the models for the maximum flood level
data set.

Models −`ˆ AIC W A KS p-Value

MKw −15.7652 −27.5305 0.0437 0.2959 0.1143 0.9562
TKw −13.6401 −21.2802 0.1420 0.8417 0.2001 0.3993
Kw −12.9732 −21.9465 0.1672 0.9746 0.2175 0.3004
Beta −14.18350 −24.3671 0.1267 0.7514 0.2062 0.3625
UR −11.1176 −20.2352 0.0892 0.5367 0.2892 0.0704
Topp −7.3813 −12.7627 0.1194 0.7110 0.3409 0.0191
Power −0.1097 1.7804 0.1229 0.7300 0.3977 0.0035
Mathematics 2021, 9, 2836 23 of 26

From Table 12, it is clear that the MKw model is the best, followed by the TKw model.
Indeed, the MKw model has the smallest AIC, W, A, KS, and the largest p-values among
all the considered models. We illustrate this claim through a graphical approach. Figure 9
plots the estimated pdfs and cdfs over the appropriate empirical objects.

5
MKw MKw
Kw Kw

1.0
Beta Beta
UR UR
4

Topp Topp

0.8
Power Power
TKw TKw
3

0.6
Density

cdf
2

0.4
0.2
1

0.0
0

0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.2 0.3 0.4 0.5 0.6 0.7 0.8

x x

(a) (b)
Figure 9. Plots of the (a) estimated pdfs and (b) estimated cdfs over the appropriate statistical objects
for the maximum flood level data set.

Based on Figure 9, we see that the fits of the MKw model are the best; the estimated
pdf has well captured the form of the histogram, and well detected the lack of observations
over the interval [0.5, 0.6]. Furthermore, the estimated cdf is very close to the scale shape of
the empirical cdf.

5.3. Air Conditioning System Data Set

The second data set involves the failure times of an airplane’s air cooling system (in
hours), as reported by [28]. The data are 23, 261, 87, 7, 120, 14, 62, 47, 225, 71, 246, 21, 42, 20,
5, 12, 120, 11, 3, 14, 71, 11, 14, 11, 16, 90, 1, 16, 52, 95. We perform a normalization operation
on these data by dividing them by 265 to obtain data between 0 and 1. To put it another
way, we work with the following data: 0.086792453, 0.984905660, 0.328301887, 0.026415094,
0.452830189, 0.052830189, 0.233962264, 0.177358491, 0.849056604, 0.267924528, 0.928301887,
0.079245283, 0.158490566, 0.075471698, 0.018867925, 0.045283019, 0.452830189, 0.041509434,
0.011320755, 0.052830189, 0.267924528, 0.041509434, 0.052830189, 0.041509434, 0.060377358,
0.339622642, 0.003773585, 0.060377358, 0.196226415, 0.358490566. A histogram analysis
reveals that the data are decreasing and right skewed, with some data close to 1 that need
to be considered. The MKw model can theoretically fit these data due to the shape ability
of the corresponding pdf. Recently, these data were taken into account in [12] to illustrate
the applicability of the UR model. By proceeding in the same way for the maximum flood
level data set, we show that the MKw model is more appropriate for these data.
First, we determine the MLLEs and the related SEs of the considered models in
Table 13.
Table 13 shows that α̂ = 11.6229 and β̂ = 0.9719 are the MLLEs for the MKw model.
By using these estimates, we obtain the estimated pdf and cdf of the MKw model through
the substitution method.
Table 14 gives the necessary values to compare the models.
Mathematics 2021, 9, 2836 24 of 26

Table 13. MLLEs and SEs of the models for the air conditioning system data set.

Models MLLEs (SEs)

MKw 11.6229 0.9719
(α, β) (1.8815) (0.1868)
TKw 0.6274 1.1217 0.6622
(α, θ, λ) (0.1245) (0.3444) (0.2631)
Kw 0.5451 1.3837
(λ, θ ) (0.1148) (0.3361)
Beta 0.5141 1.3429
( α1 , α2 ) (0.1118) (0.3642)
UR 0.1497
(τ ) (0.0273)
Topp 0.6017
(µ) (0.1098)
Power 0.4501
(ν) (0.0821)

Table 14. Values of −`ˆ , AIC, W, A, KS and (KS p-value) of the models for the air conditioning system
data set.

Models −`ˆ AIC W A KS p-Value

MKw −17.8873 −31.7747 0.0929 0.4899 0.1261 0.7260
TKw −14.9975 −23.9950 0.1691 1.0856 0.1744 0.3210
Kw −13.5389 −23.0778 0.2109 1.3470 0.1878 0.2403
Beta −13.2463 −22.49261 0.2173 1.3858 0.1957 0.2003
UR −12.7730 −23.5461 0.1253 0.7933 0.1314 0.6776
Topp −11.9801 −21.9603 0.2379 1.5102 0.1938 0.2095
Power −12.7018 −23.4036 0.2068 1.3212 0.2031 0.1680

Table 14 clearly shows that the MKw model is the best; it has the more desirable AIC,
W, A, KS, and p-values of all the models studied. We use a graphical way to demonstrate
this assertion. The estimated pdfs and cdfs are plotted over the corresponding empirical
objects in Figure 10.
5

MKw MKw
Kw Kw
1.0

Beta Beta
UR UR
4

Topp Topp
0.8

Power Power
TKw TKw
3

0.6
Density

cdf
2

0.4
0.2
1

0.0
0

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

x x

(a) (b)
Figure 10. Plots of the (a) estimated pdfs and (b) estimated cdfs over the appropriate statistical
objects for the air conditioning system data set.

According to Figure 10, the MKw model fits best; the predicted pdf captures the form
of the histogram quite well and has taken into account the values close to 1, contrary to the
other models. Furthermore, the scale shape of the empirical cdf is extremely similar to the
scale shape of the estimated cdf.
Mathematics 2021, 9, 2836 25 of 26

6. Conclusions
In this paper, a new distribution modifying the functional capabilities and mathe-
matical structure of the Kumaraswamy distribution is proposed. We have called it the
modified Kumaraswamy distribution. Evidence shows that it is more efficient than the
Kumaraswamy distribution for the modeling of phenomena with data presenting a his-
togram that is monotonic in a nearly angular way, “strongly” left- or right-skewed, and of
a leptokurtic nature, or N-shaped. On the stochastic plan, the modified Kumaraswamy dis-
tribution first order stochastic dominates the Kumaraswamy distribution. It is also deeply
connected with the so-called ratio power-logarithmic distribution. In the first part, we have
exhibited its main quantile and moment properties, with mathematical results, numerical
tables, and graphics. These elements have offered the necessary comprehension of the dis-
tribution for further theoretical and practical purposes. In the statistical plan, an inferential
methodology of the related model is developed. Then, two different data sets are analyzed
with the proposed model, and other models of references: the Kumaraswamy, beta, unit
Rayleigh, Topp–Leone and power models. The results are quite favorable to our modeling
strategy. This study is thus encouraging for the use of the modified Kumaraswamy model
for other statistical applications of importance in fields in full expansion, such as medicine,
finance, biology, and environmental sciences. The development of likelihood inferential
methods that take into account censored data, extensions to the multivariate case, incorpo-
ration of time series, spatial, and quantile regression structures in the modeling, and the
development of influence diagnostic tools are some future research directions.

Author Contributions: Conceptualization, R.A.R.B., C.C., F.J., M.E., W.A. and A.A.A.; methodology,
R.A.R.B., C.C., F.J., M.E., W.A. and A.A.A.; validation, R.A.R.B., C.C., F.J., M.E., W.A. and A.A.A.;
writing—review and editing, R.A.R.B., C.C., F.J., M.E., W.A. and A.A.A. All authors have read and
agreed to the published version of the manuscript.
Funding: This work was funded by the Deanship of Scientific Research (DSR), King AbdulAziz
University, Jeddah, under Grant No. FP-041-43.
Data Availability Statement: The corresponding author can provide the data sets utilized in this
work upon reasonable request.
Acknowledgments: The authors thank the three reviewers for the deep and constructive comments
on the first version of the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Gupta, P.L.; Gupta, R.C. The monotonicity of the reliability measures of the beta distribution. Appl. Math. Lett. 2000, 13, 5–9.
[CrossRef]
2. Ghitany, M.E. The monotonicity of the reliability measures of the beta distribution. Appl. Math. Lett. 2004, 17, 1277–1283.
[CrossRef]
3. Ferrari, S.L.; Cribari-Neto, F. Beta regression for modelling rates and proportions. J. Appl. Stat. 2004, 31, 799–815. [CrossRef]
4. Kumaraswamy, P. A generalized probability density function for double-bounded random processes. J. Hydrol. 1980, 46, 79–88.
[CrossRef]
5. Jones, M.C. Kumaraswamy’s distribution: A beta-type distribution with some tractability advantages. Stat. Methodol. 2009, 6,
70–81. [CrossRef]
6. Khan, M.S.; King, R.; Hudson, I. Transmuted Kumaraswamy distribution. Stat. Transit. New Ser. 2016, 17, 1–28.
7. Sharma, D.; Chakrabarty, T. On Size Biased Kumaraswamy Distribution. Statistics. Optim. Inf. Comput. 2016, 4, 252–264.
8. Altun, E.; Hamedani, G.G. The log-xgamma distribution with inference and application. J. Soc. Fr. Stat. 2018, 159, 40–55.
9. Mazucheli, J.; Menezes, A.F.; Dey, S. Unit-Gompertz distribution with applications. Statistica 2019, 79, 25–43.
10. Gündüz, S.; Korkmaz, M.Ç. A new unit distribution based on the unbounded Johnson distribution rule: The unit Johnson SU
distribution. Pak. J. Stat. Oper. Res. 2020, 16, 471–490. [CrossRef]
11. Altun, E. The log-weighted exponential regression model: Alternative to the beta regression model. Commun.-Stat.-Theory Methods
2020. [CrossRef]
12. Bantan, R.A.R.; Chesneau, C.; Jamal, F.; Elgarhy, M.; Tahir, M.H.; Aqib, A.; Zubair, M.; Anam, S. Some new facts about the
unit-Rayleigh distribution with applications. Mathematics 2020, 8, 1954. [CrossRef]
Mathematics 2021, 9, 2836 26 of 26

13. Haq, M.A.; Hashmi, S.; Aidi, K.; Ramos, P.L.; Louzada, F. Unit modified Burr-III distribution: Estimation, characterizations and
validation test. Ann. Data Sci. 2020. [CrossRef]
14. Korkmaz, M.Ç.; Chesneau, C.; Korkmaz, Z.S. On the arcsecant hyperbolic normal distribution. Properties, quantile regression
modeling and applications. Symmetry 2021, 13, 117. [CrossRef]
15. Korkmaz, M.Ç.; Chesneau, C. On the unit Burr-XII distribution with the quantile regression modeling and applications. Comput.
Appl. Math. 2021, 40, 1–26. [CrossRef]
16. Chesneau, C. A note on an extreme left skewed unit distribution: Theory, modelling and data fitting. Open Stat. 2021, 2, 1–23.
[CrossRef]
17. Korkmaz, M.Ç.; Chesneau, C.; Korkmaz, Z.S. Transmuted unit Rayleigh quantile regression model: Alternative to beta and
Kumaraswamy quantile regression models. Univ. Politeh. Buchar. Sci. Bull.-Ser. A-Appl. Math. Phys. 2021, 83, 149–158.
18. Figueroa-Zu, J.I.; Niklitschek-Soto, S.A.; Leiva, V.; Liu, S. Modeling heavy-tailed bounded data by the trapezoidal beta dis-
tribution with applications. REVSTAT-Stat. J. 2021, Revstat, to appear. Available online: https://www.ine.pt/revstat/pdf/
ModelingBoundedDataWithHeavyTails.pdf (accessed on 4 November 2021).
19. Bakouch, H.S.; Nik, A.S.; Asgharzadeh, A.; Salinas, H.S. A flexible probability model for proportion data: Unit-half-normal
distribution, Communications in Statistics-Case Studies. Data Anal. Appl. 2021. [CrossRef]
20. Chesneau, C. Study of a unit power-logarithmic distribution. Open J. Math. Sci. 2021, 5, 218–235. [CrossRef]
21. Corless, R.M.; Gonnet, G.H.; Hare, D.E.G.; Jeffrey, D.J.; Knuth, D.E. On the Lambert W function. Adv. Comput. Math. 1996, 5,
329–359. [CrossRef]
22. Gilchrist, W. Statistical Modelling with Quantile Functions; CRC Press: Abingdon, UK, 2000.
23. Casella, G.; Berger, R.L. Statistical Inference; Brooks/Cole Publishing Company: Bel Air, CA, USA, 1990.
24. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria,
2013.
25. Topp, C.W.; Leone, F.C. A family of J-shaped frequency functions. J. Am. Stat. Assoc. 1955, 50, 209–219. [CrossRef]
26. Konishi, S.; Kitagawa, G. Information Criteria and Statistical Modeling; Springer: New York, NY, USA, 2007.
27. Dumonceaux, R.; Antle, C.E. Discrimination between the Log-Normal and the Weibull distributions. Technometrics 1973, 15,
923–926. [CrossRef]
28. Linhart, H.; Zucchini, W. Model Selection; Wiley: New York, NY, USA, 1986.