See discussions, stats, and author profiles for this publication at: [Link]

net/publication/375553360

Characterizations of Continuous Uniform Distribution by Some Functional

Relations of Characteristic Functions

Article · January 2024

DOI: 10.18576/jsapl/110105

CITATIONS READS

0 35

2 authors:

Mohammad Ahsanullah M. SHAKIL

Rider University Miami Dade College
375 PUBLICATIONS 3,924 CITATIONS 156 PUBLICATIONS 951 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by M. SHAKIL on 10 November 2023.

The user has requested enhancement of the downloaded file.

J. Stat. Appl. Pro. Lett. 11, No. 1, 33-40 (2024) 33
Journal of Statistics Applications & Probability Letters
An International Journal

[Link]

Characterizations of Continuous Uniform Distribution

by Some Functional Relations of Characteristic Functions
M. Ahsanullah1 and M. Shakil2,*
1
Rider University, Lawrenceville, NJ, USA
2
Department of Mathematics, Miami Dade College, Hialeah, FL, USA

Received: 18 Sep. 2023, Revised: 26 Oct. 2023, Accepted: 4 Nov. 2023

Published online: 1 Jan. 2024
Abstract: Characteristic function is an important property of statistical distribution. The characteristic function uniquely
determines a probability distribution. In this article, we have characterized the continuous uniform distribution by the
characteristic function and observed that these characterizations may serve as a basis for possible applications in parameter
estimation, goodness-of-fit tests, efficiency of a particular hypothesis test, and other applied problems of real life.
Mathematics Subject Classification: 60E05, 60E10, 62E10, 62E15
Keywords: Characteristic function, characterizations, uniform distribution.

1 Introduction
Many authors and researchers have investigated the characterizations of continuous probability distributions. Before we
fit a particular probability distribution to the real-world data, it becomes necessary to justify whether the given probability
distribution satisfies the underlying requirements by its characterization. For details on characterizations of continuous
probability distributions, the interested readers are referred to Ahsanullah [1], Ahsanullah et al. [2, 3, 4], Galambos and
Kotz [5], Kotz and Shanbhag [6], and Nagaraja [7], among others. The characterization of the uniform distribution is also
one of the most important research areas of research of both pure and applied problems. In recent years, many researchers
have studied the characterizations of the uniform distribution. See, for example, Ahsanullah [8, 9], Hossain and Ahsanullah
[10], Arslan et al. [11], Hamedani and Volkmer [12], Arnold et al. [13], Arslan [14], Huang et al. [15], Nadarajah et al.
[16], and Lee [17], among others. In this paper, we have considered the characterization of the uniform distribution,
𝑋 ~𝑈(𝑎, 𝑏), defined for a positive continuous random variable X , by the characteristic function. Since the
characterization of a particular probability distribution states that it is the only distribution that satisfies some specified
conditions, these characterizations may serve as a basis for parameter estimation, see Glänzel [18, 19] and Glänzel et al.
[20]. According to Glänzel [19], the characterizations may also be useful in developing some goodness-of-fit tests of
distributions by using data whether they satisfy certain properties given in the characterizations of distributions. These
conditions are used by various authors to test goodness of fit, efficiency of a particular test of hypothesis and the power
of a particular estimating, etc. As pointed out by Volkova et al. [21], “At present, tests based on characterizations and
efficiency of such tests are not well studied; this field contains relatively few publications”. For details, the interested
readers are referred to Volkova et al. [21] and the references therein. The general problem of testing uniformity is stated
as follows: Let 𝑋) , 𝑋* , . . . , 𝑋, be independent, identically distributed random variables (i.i.d.r.v.’s) having a continuous
distribution function (d.f.) 𝐹. Then, the test of the uniformity hypothesis is given by 𝐻/ : 𝐹(𝑥) = 𝑥, 𝑥 ∈ [0,1]. For example,
Volkova et al. [21] used a well-known characterization result of Ahsanullah [6] to test the uniformity of a distribution.
Similarly, Volkova and Nikitin [22] used a well-known characterization result of Ahsanullah [23] to test exponentiality
of a distribution. As such, motivated by the importance of and possible applications of the characterizations of probability
distributions in parameter estimation and goodness-of-fit tests, and other applied problems of real life, in this paper we
have established some new characterization results of the uniform distribution, 𝑋~𝑈(𝑎, 𝑏), by the characteristic function.

The organization of this paper is as follows: In Section 2, the characteristic function of a continuous probability

*Corresponding author e-mail: mshakil@[Link]

© 2024 NSP
Natural Sciences Publishing Cor.
34 M. Ahsanullah, M. Shakil: Characterizations of Continuous Uniform …

distribution is defined, and some of its basic essential properties for the purpose of our proposed characterizations are
presented. Based on these properties, we have established some new characterization results of the uniform distribution
by the characteristic function in Section 3. The concluding remarks are manifested in Section 4.

2. Characteristic Function and Some Properties of the Uniform Distribution

In this section, we first briefly discuss the characteristic function of a continuous probability distribution and its
properties. For details on the characteristic function, the interested readers are referred to Polya [24], Lukacs and Laha
[25], Ramchandran [26], Lukacs [27, 28], and Askey [29], among others.
Definition 2.1. Characteristic Function:
The characteristic function (cf), 𝛷: (𝑡) of a continuous probability density function (pdf), (𝑥), − ∞ < 𝑥 < ∞, is defined
as
F
𝛷: (𝑡) = 𝐸(𝑒 AB: ) = ∫GF 𝑒 ABD 𝑑𝐹(𝑥) = ∫GF 𝑒 ABD 𝑓(𝑥)𝑑𝑥,
F i = -1 .

Properties:
(i) The cf, 𝛷: (𝑡) of a pdf, 𝑓(𝑥), − ∞ < 𝑥 < ∞, is a real valued function that uniquely determines the probability
distribution.
(ii) If the characteristic function of a random variable is integrable, then its cumulative distribution function (cdf), 𝐹(𝑥)
is absolutely continuous and its pdf, 𝑓(𝑥), is given by
) F
𝑓(𝑥) = *I ∫GF 𝑒 G ABD 𝛷: (𝑡) 𝑑𝑡.
(iii) A convex combination of a finite or countable number of characteristic functions is also a characteristic function.

(iv) The product of a finite number of characteristic functions is also a characteristic function.

(v) Polya [24] showed that if, for a non-negative continuous random variable X , the cf, 𝛷: (𝑡) is such that

(a) 𝛷: (0) = 1,

(b) 𝛷: (𝑡) is convex for 𝑡 > 0, and

(c) 𝛷: (∞) = 0,

then 𝛷: (𝑡) is the cf of an absolutely continuous cdf, 𝐹(𝑥) symmetric about 0.

(vi) The cf can also be used to find the moments. If a random variable, 𝑋, has moments up to the 𝑛 (> 0)𝑡ℎ order, then
the cf, 𝛷: (𝑡), is 𝑛 times continuously differentiable on − ∞ < 𝑥 < ∞, and

N O PQ (B)
𝐸(𝑋 , ) = 𝑖 G , NB O
R .
BS/
(vii) Using the definition of the moment generating function (MGF), the cf, 𝛷: (𝑡), is also given by
𝛷: (𝑡) = 𝑀: (𝑖𝑡).

3. Characterizations of the Uniform Distribution by Characteristic Functions

In this section, we present the proposed characterizations of the continuous uniform distribution, 𝑋~𝑈(𝑎, 𝑏), by its
characteristic function. For the sake of completeness, the definition and some basic essential distributional properties of
the uniform distribution are also briefly provided in this section. For details on the uniform distribution, see, for
example, Patel et al. [30], Johnson et al. [31], Balakrishnan and Nevzorov [32], and Forbes et al. [33], among others.

3.1. Uniform Distribution:

Definition: A continuous random variable 𝑋 is said to have a uniform distribution over the interval [𝑎, 𝑏], denoted
as 𝑋~𝑈(𝑎, 𝑏), if its pdf, 𝑓: (𝑥), and cumulative distribution function, 𝐹: (𝑥), are respectively given by

© 2024 NSP
Natural Sciences Publishing Cor.
J. Stat. Appl. Pro. Lett. 11, No. 1, 1-5 (2024) / [Link] 35
)
VGW
,𝑎 ≤ 𝑥 ≤ 𝑏
𝑓: (𝑥) = U ,
0, 𝑥 < 𝑎 𝑜𝑟 𝑥 > 𝑏

and
0, < 𝑎
DGW
𝐹: (𝑥) = UV G W , 𝑎 ≤ 𝑥 ≤ 𝑏 ,
1, 𝑥 > 𝑏
where 𝑎 is the location parameter, 𝑏 − 𝑎 s the scale parameter, and − ∞ < 𝑎 < 𝑏 < ∞.
When 𝑎 = 0 and 𝑏 = 1, it is called the standard uniform distribution, with its pdf given by

𝑓: (𝑥) = 1, 𝑓𝑜𝑟 0 ≤ 𝑥 ≤ 1.
Properties:
Moments: Some of these are given as follows: \

VO [ \ G WO [ \
(i) 𝑛𝑡ℎ Moment: 𝐸(𝑋 , ) = (, ] )) (V G W)

W]V
(ii) Expected Value (or 1st Moment): 𝐸(𝑋) =
*

(V G W)_
(i) Variance (or 2nd Central Moment): 𝑉(𝑋) =
)*

a bc G a bd
B (V G W)
, 𝑡 ≠0
Moment Generating Function: 𝑀: (𝑡) = ` .
1, 𝑡= 0
a f bc G a f bd
, 𝑡 ≠0
A B (V G W)
Characteristic Function: 𝛷: (𝑡) = 𝐸(𝑒 AB: ) = 𝑀: (𝑖𝑡) = ` .
1, 𝑡= 0
3.2. Characterizations
In what follows, we establish the proposed characterization results of the continuous uniform distribution by the
characteristic function. Klebanov [34] proved that if 𝜑(𝑡) is the characteristic function of an absolutely continues
random variable symmetric about 0, then the relation
(1/2)(𝜑(𝑡) + 𝜑(𝑡/2))(𝜑(𝑡/4))² = (𝜑(𝑡/2))³

characterizes a uniform distribution. We will present here a slight difference of the Klebanov’s theorem [34] and an
alternative proof.

3.2.1. Main Results

[Link]. Results Based on the Characteristic Function

Theorem 3.1. Suppose a random variable 𝑋 is absolutely continuous with cdf, 𝐹(𝑥), where 𝐹(−1) = 0 and 𝐹(1) = 1.
Suppose 𝑋 has a uniform distribution symmetric about 0 and 𝜑(𝑡) is the cf of 𝑋, that is,

𝛷: (𝑡) = 𝐸(𝑒 AB: ) = ∫G)

) a fbr
𝑑x =
a fb G a tfb
=
uA, B
, i = -1 .
* *AB B

Then
(1/2)(𝜑(𝑡) + 𝜑(𝑡/2))(𝜑(𝑡/4))² = (𝜑(𝑡/2))³

if and only if

𝐹(𝑥) = ((𝑥 + 1)/2), −1 ≤ 𝑥 ≤ 1,

© 2024 NSP
Natural Sciences Publishing Cor.
36 M. Ahsanullah, M. Shakil: Characterizations of Continuous Uniform …

which is a uniform distribution symmetric about 0, with its cf as given above.

Proof. Suppose that the random variable 𝑋 has a uniform distribution symmetric about 0. Then its cf is given by

𝜑(𝑡) = (𝑠𝑖𝑛 𝑡)/𝑡. (3.1)

B
Replacing 𝑡 by * in the equation (3.1), we have

B B
𝜑(*) = 2(𝑠𝑖𝑛 *)/𝑡. (3.2)

Let 𝑚(𝑡/2) = (𝜑(𝑡)/𝜑(𝑡/2). From the equations (3.1) and (3.2), it is easily seen that

𝑚(𝑡/2) = ((𝜑(𝑡))/(𝜑(𝑡/2))) = 𝑐𝑜𝑠 (𝑡/2). (3.3)

B
Replacing 𝑡 by * in the equation (3.3), we have

𝑚(𝑡/4) = ((𝜑(𝑡/2))/(𝜑(𝑡/4))) = 𝑐𝑜𝑠 (𝑡/4). (3.4)

From the equations (3.3) and (3.4), and using the trigonometric identity, 𝑐𝑜𝑠 (2𝛳) = 2 𝑐𝑜𝑠 * 𝛳 − 1, it is easily seen
that
)
(𝑚(𝑡/2) + 1) = (𝑚(𝑡/4))* . (3.5)
*

Thus, in view of the equations (3.3) and (3.4), the equation (3.5) easily reduces to

(1/2)(𝜑(𝑡) + 𝜑(𝑡/2))(𝜑(𝑡/4))² = (𝜑(𝑡/2))³ .

This proves the “if condition”.

Now, we prove the “only if condition”. Suppose that the relation

(1/2)(𝜑(𝑡) + 𝜑(𝑡/2))(𝜑(𝑡/4))² = (𝜑(𝑡/2))³ . (3.6)

holds. Dividing both sides of the equation (3.6) by (𝜑(𝑡/4))², and substituting 𝑚(𝑡/2) = (𝜑(𝑡)/𝜑(𝑡/2), we obtain
)
*
(𝑚(𝑡/2) + 1) = (𝑚(𝑡/4))* . (3.7)

From the equations (3.3) and (3.4), and using the trigonometric identity, 𝑐𝑜𝑠 (2𝛳) = 2 𝑐𝑜𝑠 * 𝛳 − 1, it is easily seen
that the solution of the equation (3.7) is given by

𝑚(𝑡/2) = 𝑐𝑜𝑠 (𝑡/2). (3.8)

Now, using (3.8), we have

φ(t) = 𝑚(𝑡/2)𝜑(𝑡/2) = 𝑐𝑜𝑠 (𝑡/2) 𝜑(𝑡/2) = 𝑐𝑜𝑠 (𝑡/2) 𝑐𝑜𝑠 (𝑡/4) 𝜑(𝑡/4) = . .. ,

uA, | F |
or, proceeding in the same manner, and using the formula 38.8, = Π~S) cos ‚ ƒ„ , page 207, Spiegel et al. [35],
| *
and 𝜑(0) = 1, we have

F B uA,B
φ(t) = Π~S) cos ‚ ƒ„ 𝜑(0) = ,
* B

which is the required cf of the uniform distribution on [−1,1]. This completes the proof of the Theorem 3.1.

[Link]. Results Based on Functional Relations of the Characteristic Functions

© 2024 NSP
Natural Sciences Publishing Cor.
J. Stat. Appl. Pro. Lett. 11, No. 1, 1-5 (2024) / [Link] 37

We will present here two theorems using functional relations of the characteristic functions.

Theorem 3.2. Suppose a random variable 𝑋 has an absolutely continuous cdf, 𝐹(𝑥), with 𝐹(−1) = 0 and 𝐹(1) = 1.
We assume the random variable 𝑋 is symmetric about 0. Let the corresponding cf be 𝜑(𝑡).Then
𝑡𝜑′′(𝑡) + 2𝜑′(𝑡) + 𝑡𝜑(𝑡) = 0,
if and only if

𝐹(𝑥) = ((𝑥 + 1)/2), −1 ≤ 𝑥 ≤ 1.

which is a uniform distribution symmetric about 0, with its cf as defined in Theorem 3.1.

Proof. For the uniform distribution on [−1, 1], we have its cf given by

𝜑(𝑡) = (𝑠𝑖𝑛 𝑡)/𝑡. (3.9)

It is easy to see, from the equation (3.9), that

𝑡𝜑′′(𝑡) + 2𝜑′(𝑡) + 𝑡𝜑(𝑡) = 0.

This proves the “if condition”.

Now, we shall prove the “only if condition”. Suppose that

𝑡𝜑′′(𝑡) + 2𝜑′(𝑡) + 𝑡𝜑(𝑡) = 0. (3.10)

We can write the equation (3.10) as

𝑡𝜑′′(𝑡) + 𝜑′(𝑡) + 𝜑′(𝑡) = −𝑡𝜑(𝑡).

That is,

((𝑑²)/(𝑑𝑡²))(𝑡𝜑(𝑡)) = −𝑡𝜑(𝑡). (3.11)

Let

𝛹(𝑡) = 𝑡𝜑(𝑡). (3.12)

Then, substituting (3.12) in the equation (3.11), we have

𝛹′′(𝑡) + 𝛹(𝑡) = 0 . (3.13)

The solution of the (3.13) is easily given by

𝛹(𝑡) = 𝑎𝑒 AB + 𝑏𝑒 GAB , (3.14)

where 𝑎 and 𝑏 are arbitrary constants. Now, from the equation (3.14), using the definition of the cf, we have 𝛹(0) = 0
and 𝛹′(0) = 1. Using these conditions in the equation (3.14) and solving for 𝑎 and 𝑏, we must have

a fb G a tfb
𝛹(𝑡) = = 𝑠𝑖𝑛 𝑡. (3.15)
*A

Thus, from the equations (3.12) and (3.15), it follows that

𝜑(𝑡) = (𝑠𝑖𝑛 𝑡)/𝑡,

which is the cf of the uniform distribution function on [−1, 1]. This completes the proof of the Theorem 3.2.

Theorem 3.3. Suppose a random variable 𝑋 has an absolutely continuous pdf with cdf, 𝐹(𝑥), where 𝐹(−1) = 0 and
𝐹(1) = 1. We assume the random variable 𝑋 is symmetric about 0. Let the corresponding cf be 𝜑(𝑡).Then
© 2024 NSP
Natural Sciences Publishing Cor.
38 M. Ahsanullah, M. Shakil: Characterizations of Continuous Uniform …

(𝜑(2𝑡))² = (𝜑(𝑡))² − 𝑡²(𝜑(𝑡))⁴ ,

if and only if

𝐹(𝑥) = ((𝑥 + 1)/2), −1 ≤ 𝑥 ≤ 1.

Proof. It is easy to prove the necessary condition. We will prove here the sufficient condition. Suppose that

(𝜑(2𝑡))² = (𝜑(𝑡))² − 𝑡²(𝜑(𝑡))⁴. (3.16)

Let
𝑡𝜑(𝑡) = 𝛹(𝑡). (3.17)

Then, from the equations (3.17) and (3.18), after simplifications, we easily have

(𝛹(2𝑡))² = 4(𝛹(𝑡))² − 4(𝛹(𝑡))⁴. (3.18)

Following the similar arguments as in the Theorem 3.2, the solution of the equation (3.18) under the condition 𝛹(0) =
0 and 𝛹′(0) = 1 is easily given by

𝛹(𝑡) = 𝑎𝑒 AB + 𝑏𝑒 GAB . (3.19)

Since 𝛹(0) = 0 and 𝛹′(0) = 1, Using these conditions in the equation (3.19) and solving for 𝑎 and 𝑏, we must have

a fb G a tfb
𝛹(𝑡) = = 𝑠𝑖𝑛 𝑡. (3.20)
*A

Thus, from the equations (3.17) and (3.20), it follows that

𝜑(𝑡) = (𝑠𝑖𝑛 𝑡)/𝑡,

which is the cf of the uniform distribution function on [−1, 1]. This completes the proof of the Theorem 3.3.

4. Concluding Remarks
As stated in the introduction, the characterization of a particular probability distribution states that it is the only
distribution that satisfies some specified conditions, and hence these characterizations may serve as a basis for parameter
estimation. As such, motivated by the importance of and possible applications of the characterizations of probability
distributions in parameter estimation and goodness-of-fit tests, and other applied problems of real life, in this paper, we
have established some new characterization results of the uniform distribution, 𝑋~𝑈(𝑎, 𝑏), by the characteristic function
based on some of its basic essential distributional properties discussed in the paper. We sincerely believe that this paper
will be further helpful to test the uniformity of a distribution analogously to Volkova and Nikitin (2013) and Volkova et
al. (2020). Moreover, we hope that the paper will be quite useful for other researchers and practitioners in the fields of
probability, statistics, and other applied sciences.

Acknowledgements
The authors are thankful to the editor-in-chief and the three anonymous reviewers whose constructive comments and
suggestions have improved the quality and presentation of the paper.
Authors’ Contributions: All authors have equally made contributions. All authors read and approved the final
manuscript.
Competing Interests: The authors declare that they have no competing interests.
Funding: The authors state that they have no funding source for this paper.
Availability of Data and Materials: Not applicable.
© 2024 NSP
Natural Sciences Publishing Cor.
J. Stat. Appl. Pro. Lett. 11, No. 1, 1-5 (2024) / [Link] 39

Declarations Conflict of interest: The authors declare that they have no competing interests. The authors state that no
funding source or sponsor has participated in the realization of this work.
References:
[1] M. Ahsanullah, Characterizations of Univariate Continuous Distributions, Atlantis-Press, Paris, France (2017).
[2] M. Ahsanullah, B. M. G. Kibria and M. Shakil, Normal and Student´s t Distributions and Their Applications,
Atlantis Press, Paris, France (2014).
[3] M. Ahsanullah, M. Shakil and B. M. G. Kibria, Characterizations of folded student’s t distribution, Journal of
Statistical Distributions and Applications, 2(1), 15 (2015).
[4] M. Ahsanullah, M. Shakil and B. M. G. Kibria, Characterizations of Continuous Distributions by Truncated
Moment, Journal of Modern Applied Statistical Methods, 15(1), 17 (2016).
[5] J. Galambos and S. Kotz, Characterizations of probability distributions: A unified approach with an emphasis on
exponential and related models (Lecture Notes in Mathematics), 675, Springer, Berlin, Germany (1978).
[6] S. Kotz and D. N. Shanbhag, Some new approaches to probability distributions, Advances in Applied Probability, 12,
903 – 921 (1980).
[7] H. Nagaraja, Characterizations of Probability Distributions, In: H. Pham (eds) Springer Handbook of Engineering
Statistics, Springer Handbooks, Springer, London, UK, 79 – 95, (2006). DOI: [Link]
1_4
[8] M. Ahsanullah, A Characterization of the Uniform Distribution by Dual Generalized Order
Statistics, Communications in Statistics - Theory and Methods, 33:12, 2921-2928 (2004). DOI: 10.1081/STA-
200038854
[9] M. Ahsanullah, On characterizations of the uniform distribution based on functions of order statistics, Aligarh J.
Statist., 9, 1–6 (1989).
[10] S. A. Hossain, and M. Ahsanullah, On Characterizations of Distributions by Truncated Moments, 69(1), 65-71
(2012)
[11] G. Arslan, M. Ahsanullah and I. G. B. Bairamov, On Characteristic Properties of the Uniform
Distribution, Sankhyā: The Indian Journal of Statistics (2003-2007), 67(4), 715–721 (2005). DOI:
[Link]
[12] G. Hamedani and H. Volkmer, Certain characterizations of the uniform distribution. Metrika, 61, 117–125 (2005).
DOI: [Link]
[13] B. C. Arnold, E. Castillo and J. Sarabia, Some characterizations involving uniform and powers of uniform random
variables. Statistics, 42, 527-534 (2008). DOI: 10.1080/02331880801987836.
[14] G. Arslan, On a characterization of the uniform distribution by generalized order statistics. Journal of
Computational and Applied Mathematics, 235(16), 4532-4536 (2011). DOI: [Link]
[15] J. S. Huang, B. C. Arnold and M. Ghosh, On Characterizations of the Uniform Distribution Based on Identically
Distributed Spacings. Sankhyā: The Indian Journal of Statistics, Series B (1960-2002), 41(1/2), 109–115 (1979). DOI:
[Link]
[16] S. Nadarajah, M. Teimouri and S. H. Shih, Characterizations of the Weibull and uniform distributions using record
values, Braz. J. Probab. Stat., 28(2), 209 – 222 (2014). DOI: [Link]
[17] M-Y. Lee, Characterizations of the uniform distributions based on upper record values, J. Appl. Math. &
Informatics, 37(5 – 6), 351 – 355 (2019). DOI: [Link]
[18] W. Glänzel, A characterization theorem based on truncated moments and its application to some distribution
families, In Mathematical Statistics and Probability Theory, Springer, Dordrecht, Germany, (75-84), (1987).
[19] W. Glanzel, Some consequences of a characterization theorem based on truncated moments, Statistics, 21(4), 613-
618 (1990).
[20] W. Glänzel, A. Teles and A. Schubert, Characterization by truncated moments and its application to Pearson-type
distributions, Zeitschrift für Wahrscheinlichkeitstheorie und verwandte Gebiete, 66(2), 173-183 (1984).
[21] K. Y. Volkova, M. S. Karakulov and Y. Y. Nikitin, Goodness-of-Fit Tests Based on Characterization of Uniformity
by the Ratio of Order Statistics, and Their Efficiency, J Math Sci, 244, 743–751 (2020). DOI:
[Link]
[22] K. Y. Volkova and Y. Y. Nikitin, Exponentiality tests based on Ahsanullah’s characterization and their efficiency,
Journal of Mathematical Sciences, 204(1), 42-54 (2015).
[23] M. Ahsanullah, A characterization of the exponential distribution by spacings, Journal of Applied Probability,
15(3), 650-653 (1978).
[24] G. Pólya, Remarks on characteristic functions, in Proc. First Berkeley Conf. on Math. Stat. and Prob., (115-123),
(1949).
[25] E. Lukacs and R. G. Laha, Applications of Characteristic Functions, Griffin, London, UK (1964).
© 2024 NSP
Natural Sciences Publishing Cor.
40 M. Ahsanullah, M. Shakil: Characterizations of Continuous Uniform …

[26] B. Ramachandran, Advanced Theory of Characteristic Functions, Statistical Publishing Society, Calcutta, India
(1967).
[27] E. Lukacs, Characteristic Functions, 2nd rev. Ed., Griffin, London, UK (1970).
[28] E. Lukacs, A Survey of the Theory of Characteristic Functions, Advances in Applied Probability, 4(1), 1–38 (1972).
DOI: [Link]
[29] R. Askey, Some characteristic functions of unimodal distributions. Journal of Mathematical Analysis and
Applications, 50(3), 465-469 (1975). DOI: [Link]
[30] N. Balakrishnan and V. B. Nevzorov, A Primer on Statistical Distributions, John Wiley & Sons, Inc., New York,
USA (2003).
[31] C. Forbes, M. Evans, N. Hastings and B. Peacock, Statistical Distributions, 4th Edition, John Wiley & Sons, Inc.,
New York, USA (2010).
[32] N. L. Johnson, S. Kotz, and N. Balakrishnan, Continuous Univariate Distributions, Volumes 1 – 2, 6th Edition.
John Wiley & Sons, Inc., New York, USA (1994).
[33] J. K. Patel, C. H. Kapadia and D. B. Owen, Handbook of Statistical Distributions, Marcel Dekker, New York, USA
(1976).
[34] L. B. Klebanov, Characterizations of Probability Distributions by the Properties of Linear Forms with Random
Coefficients, J. Math. Sci., 273, 778–786 (2023). DOI: [Link]
[35] M. R. Spiegel, S. Lipschutz and J. Liu, Mathematical Handbook of Formulas and Tables, McGraw-Hill Co., New
York, USA (2009).

© 2024 NSP
Natural Sciences Publishing Cor.

View publication stats