Next Article in Journal
Mathematical Entropy Analysis of Natural Convection of MWCNT—Fe3O4/Water Hybrid Nanofluid with Parallel Magnetic Field via Galerkin Finite Element Process
Next Article in Special Issue
Modeling the Amount of Carbon Dioxide Emissions Application: New Modified Alpha Power Weibull-X Family of Distributions
Previous Article in Journal
Sandstorm Image Enhancement Using Image-Adaptive Eigenvalue and Brightness-Adaptive Dark Channel Network
Previous Article in Special Issue
An Improved Equilibrium Optimizer with a Decreasing Equilibrium Pool
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Arcsine Kumaraswamy-Generalized Family: Bayesian and Classical Estimates and Application

Department of Statistics and Operations Research, Faculty of Science, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia
*
Author to whom correspondence should be addressed.
Symmetry 2022, 14(11), 2311; https://doi.org/10.3390/sym14112311
Submission received: 21 September 2022 / Revised: 23 October 2022 / Accepted: 27 October 2022 / Published: 3 November 2022

Abstract

:
In this paper, by including a trigonometric function, we propose a family of heavy-tailed distribution called the arcsine Kumaraswamy generalized-X family of distributions. Based on the proposed approach, a four-parameter extension of the Lomax distribution called the arcsine Kumaraswamy generalized Lomax (ASKUG-LOMAX) distribution is discussed in detail. Maximum likelihood, bootstrap, and Bayesian estimation are used to estimate the model parameters. A simulation study is used to evaluate ASKUG-LOMAX performance. The flexibility and usefulness of the proposed ASKUG-LOMAX distribution to predict unique symmetric and asymmetric patterns is demonstrated by analyzing real data. The results show that the ASKUG-LOMAX model is a good candidate for analyzing claims based on heavy-tailed data.

1. Introduction

In the applied sciences, distributions that are heavy-tailed play an important role in modeling data, especially in the fields of industry, economics, finance, banking, and risk management. However, the quality of these methods depends primarily on the assumed probability model of the phenomenon under consideration. In the applied fields, industrial datasets are generally positive [1], right-skewed [2], unimodal [3], and with strong outliers [4]. Right-skewed data can be adequately modeled by skewed distributions [5]. Therefore, a set of unimodal positively-skewed parametric distributions can be used to model such datasets [6,7]. When the CDF F ( u ) is an underlined distribution, the heavy-tailed models are those with probabilities at the right tail that are heavier than the exponential one (see Beirlant et al. [8] and Resnick [9]), satisfying
lim u e ( λ u ) 1 1 F ( u ) = 0 , λ > 0 .
Dutta and Perry [10] conducted an empirical analysis of loss distributions to estimate risk using various approaches. They rejected the use of the exponential, gamma, and Weibull models because of their poor results, and concluded that the best choice would be to use a model that is flexible enough in its structure. These results encouraged researchers to propose new flexible models that provide greater accuracy in data fitting. To overcome the problems associated with modeling based on classical distributions, new families of distributions (see [11,12,13,14,15,16,17]) have been introduced.
In this context, Mudholkar and Srivastava [18] have discussed the exponential family via the composition of a shape parameter to obtain a more adaptive extension of the basic model. The CDF of the random variable X over the exponential family is provided by
F u ; ζ , φ = G u ; θ ζ , ζ 0 , θ , u R ,
where G u ; φ is the CDF of the underline distribution, φ is the parameter vector, and ζ > 0 is a shape parameter. Moreover, the Kumaraswamy generalised (KUG) family is a more adaptive family proposed by Cordeiro and de Castro [19] via the composition of two shape parameters. The CDF of the random variable X over the KUG family is provided by
F u ; δ 1 , δ 2 , φ = 1 1 G u ; φ δ 1 δ 2 , δ 1 , δ 2 0 , φ , u R ,
The arcsine exponentiated X family of distributions appears here as a special case when δ 2 = 1 . It is based on Equation (3); for a contribution, see Mead and Afify [20]. With respect to Equation (3), the KUG expansion of the current distributions has been discussed in the literature; see Mansour et al. [21] and Ahmad et al. [22]. The arcsine KUG-X family is denoted by (ASKUG-X). The CDF of the random variable X over ASKUG is
F x ; δ 1 , δ 2 , φ = 2 π a r c s i n 1 1 G x ; φ δ 1 δ 2 , δ 1 , δ 2 0 , φ , x R ,
where δ 1 and δ 2 are two additional shape parameters. The main motivations for using the ASKUG-X family in practice are the following:
(i)
To develop the flexibility and properties of basic models.
(ii)
A suitable procedure for adding two extra parameters in expanded models with potent outliers, which is very useful when modeling industrial data (see Section 5).
(iii)
To introduce the extended version of a basic model with closed forms for the cdf and hazard rate function, the special submodels of this family can be used in the analysis of censored data sets.
(iv)
Compared to existing competing models, the special cases of the ASKUG-X approach are able to model data sets with high-tailed content in factorial habits.
The respective PDF, survival function (SF), and hazard rate function (HRF) of the random variable X via ASKUG are provided as follows:
f x ; δ 1 , δ 2 , φ = 2 δ 1 δ 2 g x ; φ G x ; φ δ 1 1 1 1 G x ; φ δ 1 δ 2 1 π 1 1 1 G x ; φ δ 1 δ 2 2 ,
S = 1 2 π arcSin 1 1 G x ; φ δ 1 δ 2
and
h = 2 δ 1 δ 2 g x ; φ G x ; φ δ 1 1 1 1 G x ; φ δ 1 δ 2 1 1 1 1 G x ; φ δ 1 δ 2 2 π 2 arcSin 1 1 G x ; φ δ 1 δ 2 ,
where δ 1 , δ 2 > 0 , φ R and x R . Via the new CDF of ASKUG-X, many new heavy-tailed flexible models can be obtained. A number of newly contributed models based on the ASKUG-X approach are presented in Table 1.
The rest of this paper is outlined as follows. Section 2 defines ASKUG-LOMAX. Section 3 provides the maximum likelihood estimators (MLEs). Section 4 provides the Bayesian estimators. Section 5 presents a discussion of our simulations and analyzes two examples of real data to illustrate the proposed ASKUG-LOMAX potentiality. Section 6 provides a brief conclusion.

2. ASKUG-Lomax Distribution

In this section, we introduce the ASKUG-Lomax distribution and examine the attitude of its PDF and HRF. First, we assess the CDF of the Lomax distribution, G ( x ; κ , α ) = 1 ( 1 + κ x ) α , x 0 , κ , α > 0 . Then, a random variable X is said to follow the ASKUG-Lomax distribution if its CDF takes the following form:
F ( x ; κ , α , δ 1 , δ 2 ) = 2 π arcSin 1 1 1 ( 1 + κ x ) α δ 1 δ 2 , x 0 , κ , α , δ 1 , δ 2 > 0 .
The respective PDF, SF, and HRF of the ASKUG-Lomax distribution are provided by
f ( x ; κ , α , δ 1 , δ 2 ) = 2 κ α δ 1 δ 2 1 1 + κ x α δ 1 1 1 1 1 + κ x α δ 1 δ 2 1 π 1 + κ x α + 1 1 1 1 1 1 + κ x α δ 1 δ 2 2 , x 0 , κ , α , δ 1 , δ 2 > 0 ,
S = 1 2 arcSin 1 1 1 ( 1 + x κ ) α δ 1 δ 2 π ,
and
h = 2 κ α δ 1 δ 2 ( 1 + x κ ) α 1 1 ( 1 + κ x ) α δ 1 1 1 1 ( 1 + κ x ) α δ 1 δ 2 1 π 1 1 1 1 ( 1 + κ x ) α δ 1 δ 2 2 1 2 arcSin 1 1 1 ( 1 + κ x ) α δ 1 δ 2 π .
The ASKUG-Lomax model reduces to the AS-exponentiate-Lomax distribution when δ 2 = 1 , and to the AS-Lomax distribution when δ 1 = δ 2 = 1 .

2.1. Quantile Function

Let X be the ASKUG-Lomax random variable with the PDF from Equation (5); then, the quantile function of X, i.e., Q(u), reduces to
Q ( u ) = 1 k 1 1 1 sin π u 2 1 δ 2 1 δ 1 1 α 1 .
where u has a uniform distribution on the interval (0, 1). From the expression in Equation (12), it is evident that the ASKUG-Lomax family has a closed form solution of its quantile function.

2.2. Moments

Moments are very important in statistical analysis, and play an essential role. They help to capture important features and properties of the distribution (e.g., its central tendency, dispersion, skewness, and kurtosis). The r t h moment of the ASKUG-X family is
μ r = x r f x ; δ 1 , δ 2 , φ d x .
Substituting Equation (5) into Equation (12), we obtain
μ r = x r 2 δ 1 δ 2 g x ; φ G x ; φ δ 1 1 1 1 G x ; φ δ 1 δ 2 1 π 1 1 1 G x ; φ δ 1 δ 2 2 d x .
Using the binomial expansion, we have
1 1 u 2 = t = 0 1 × 3 × 5 × × ( 2 t 1 ) 2 t × t ! u 2 t .
By replacing u with 1 G x ; φ δ 1 δ 2 in Equation (14), we obtain
μ r = 2 δ 1 δ 2 π t = 0 1 × 3 × 5 × × ( 2 t 1 ) 2 t × t ! η r , δ 1 , 2 t δ 2 ,
where η r , δ 1 , 2 t δ 2 = x r g x ; φ G x ; φ δ 1 1 1 G x ; φ δ 1 2 t δ 2 1 1 G x ; φ δ 1 δ 2 1 d x .
The moment-generating function of the ASKUG-X class has the following form:
M x ( y ) = 2 δ 1 δ 2 π t = 0 1 × 3 × 5 × × ( 2 t 1 ) 2 t × t ! × r ! y r η r , δ 1 , 2 t δ 2

3. Maximum Likelihood Estimation

Take the observed values x 1 , x 2 , , x n of X 1 , X 2 , , X n , which is a random sample from the ASKUG-X model; then, the ASKUG-X log-likelihood is
L = n Log 2 π δ 1 δ 2 + i = 1 n Log g x i ; φ + δ 1 1 i = 1 n Log G x i ; φ + i = 1 n Log 1 1 G x i ; φ δ 1 δ 2 1 1 2 i = 1 n Log 1 1 1 G x i ; φ δ 1 δ 2 2 .
The MLE can be derived by maximizing Equation (12) (see Appendix A).
The asymptotic CIs of κ , α , δ 1 and δ 2 can be computed. The variance–covariance matrix V ( κ ^ , α ^ , δ 1 ^ , δ 2 ^ ) is provided by
V ( κ , α , δ 1 , δ 2 ) = 2 L κ 2 2 L κ α 2 L κ δ 1 2 L κ δ 2 2 L α κ 2 L α 2 2 L α δ 1 2 L α δ 2 2 L δ 1 κ 2 L δ 1 α 2 L δ 1 2 2 L δ 1 δ 2 2 L δ 2 κ 2 L δ 2 α 2 L δ 2 δ 1 2 L δ 2 2 1 ,
The respective 100 ( 1 ε ) % two-sided approximate CIs for κ , α , δ 1 and δ 2 are provided by
κ ^ ± z κ / 2 V ( κ ^ ) , α ^ ± z κ / 2 V ( α ^ ) , δ 1 ^ ± z κ / 2 V ( δ 1 ^ ) , and δ 2 ^ ± z κ / 2 V ( δ 2 ^ ) ,
where V ( κ ^ ) , V ( α ^ ) , V ( δ 1 ^ ) and V ( δ 2 ^ ) are provided by the diagonal elements of V ( κ ^ , α ^ , δ 1 ^ , δ 2 ^ ) , and z ε / 2 is the upper ε 2 percentile of a standard normal distribution.
Next, to obtain the bootstrap CI boot-p for the unknown parameters ϕ = ( κ , α , δ 1 , δ 2 ), we apply the following algorithm, Algorithm 1;
Algorithm 1 Boot-p interval algorithm:
step-1:
Simulate x 1 : n , x 2 : n , , x n : n from the ASKUG-Lomax distribution and obtain an estimate ϕ ^ of ϕ .
step-2:
Simulate another sample x 1 : n * , x 2 : n * , , x n : n * via ϕ ^ . Then, obtain the updated bootstrap estimate ϕ ^ * of ϕ .
step-3:
Iterate step 2 a previously fixed number of iterations B.
step-4:
Via F ^ ( x ) = P ( ϕ ^ * x ) , that is, the CDF of ϕ ^ * , the 100 ( 1 ε ) % CI of ϕ is provided by
ϕ ^ B o o t p ( ε 2 ) , ϕ ^ B o o t p ( 1 ε 2 ) ,
where ϕ ^ B o o t p ( x ) = F ^ 1 ( x ) and x is previously fixed.

4. Bayesian Estimation

Suppose that κ , α , δ 1 , and δ 2 are random variables that follow the prior PDFs Gamma ( κ ; a 1 , b 1 ) , Gamma ( α ; a 2 , b 2 ) , Gamma ( δ 1 ; a 3 , b 3 ) , and Gamma ( δ 2 ; a 4 , b 4 ) , respectively, where a i and b i are positive constants and i = 1 , 2 , 3 , 4 . The posterior DF of κ , α , δ 1 , δ 2 and the data under the Gamma priors can take the forms
π 1 ( κ ) = b 1 a 1 Γ ( a 1 ) κ a 1 1 exp b 1 κ , κ , a 1 , b 1 > 0 ,
π 2 ( α ) = b 2 a 2 Γ ( a 2 ) α a 2 1 exp b 2 α , α , a 2 , b 2 > 0 ,
π 3 ( δ 1 ) = b 3 a 3 Γ ( a 3 ) δ 1 a 3 1 exp b 3 δ 1 , δ 1 , a 3 , b 3 > 0 ,
and
π 4 ( δ 2 ) = b 4 a 4 Γ ( a 4 ) δ 2 a 4 1 exp b 4 δ 2 , δ 2 , a 4 , b 4 > 0 .
Then, the posterior density of κ , α , δ 1 , δ 2 and the data can be extracted as
π * ( κ , α , δ 1 , δ 2 | x ) π ( κ , α , δ 1 , δ 2 ) i = 1 n f ( x i ; κ , α , δ 1 , δ 2 ) , = J 1 κ n + a 1 1 α n + a 2 1 δ 1 n + a 3 1 δ 2 n + a 4 1 e b 1 κ + b 2 α + b 3 δ 1 + b 4 δ 2 i = 1 n 1 1 + κ x i α δ 1 1 1 1 1 + κ x i α δ 1 δ 2 1 1 + κ x i α + 1 1 1 1 1 1 + κ x α δ 1 δ 2 2 ,
where x i 0 , κ , α , δ 1 , δ 2 , a i , b i > 0 , i = 1 , 2 , 3 , 4 and J is the normalizing constant.

MCMC Method

We use the Metropolis Hastings (M-H) procedure as follows:
  • Set initial values κ ( 0 ) , α ( 0 ) , δ 1 ( 0 ) , and δ 2 ( 0 ) . Then, simulate a sample of size n from ASKUG-Lomax, next set l = 1 .
  • Simulate κ ( * ) , α ( * ) , δ 1 ( * ) , and δ 2 ( * ) using the proposal distributions N ( κ ( l 1 ) , V ( κ ^ ) ) , N ( α ( l 1 ) , V ( α ^ ) ) , N ( δ 1 ( l 1 ) , V ( δ 1 ^ ) ) , and N ( δ 2 ( l 1 ) , V ( δ 2 ^ ) ) .
  • Obtain the probability r = m i n π * ( κ ( * ) , α ( * ) , δ 1 ( * ) , δ 2 ( * ) ) π * ( κ ( l 1 ) , α ( l 1 ) , δ 1 ( l 1 ) , δ 2 ( l 1 ) ) , 1 .
  • Simulate U from Uniform (0, 1).
  • If U < r , then κ ( l ) , α ( l ) , δ 1 ( l ) , δ 2 ( l ) = κ ( * ) , α ( * ) , δ 1 ( * ) , δ 2 ( * ) .
    If U r , then κ ( l 1 ) , α ( l 1 ) , δ 1 ( l 1 ) , δ 2 ( l 1 ) = κ ( * ) , α ( * ) , δ 1 ( * ) , δ 2 ( * ) .
  • Set l = l + 1 .
  • Iterate Steps 2–6, M repetitions, and obtain κ ( l ) , α ( l ) , δ 1 ( l ) and δ 2 ( l ) for l = 1 , , M .
Now, we use the squared error loss function provided by L S E ( ϑ , ϑ ^ ) = ( ϑ ϑ ^ ) 2 , where ϑ ^ is an estimate of the unknown parameter ϑ , which against the SE loss function is the posterior mean. Using the generated random samples from the above Gibbs sampling technique and with N the nburn, the Bayes estimator of ϑ , say, ϑ ^ S E , can be obtained as
ϑ ^ S E = E ϑ ϑ | x = 1 M N l = N + 1 M ϑ ( l ) .
The second loss function is the LINEX loss function, provided by
L L E ϑ , ϑ ^ = exp ρ ϑ ϑ ^ ρ ϑ ϑ ^ 1 , ρ 0 .
The approximate Bayes estimate of ϑ = σ , δ , γ under the LE loss function based on the Gibbs sampling technique becomes
ϑ ^ L E = 1 ρ log E ϑ exp ρ ϑ | x = 1 ρ log l = N + 1 M e x p ρ ϑ ( l ) M N ,
Finally, the general entropy (GE) loss function is provided by
L G E ϑ , ϑ ^ = ϑ ^ ϑ ε ε log ϑ ^ ϑ 1 .
The approximate Bayes estimate of the parameters is provided by
ϑ ^ G E = E ϑ ϑ ε x 1 ε = 1 M N l = N + 1 M ϑ ( l ) ε 1 ε ,
MCMC HPD-credible interval algorithm:
  • Sort κ ( * ) , α ( * ) , δ 1 ( * ) , and δ 2 ( * ) in rising values.
  • The lower bounds of κ , α , δ 1 , and δ 2 in the rank ( M N ) ε / 2 .
  • The lower bounds of κ , α , δ 1 , and δ 2 in the rank ( M N ) ( 1 ε / 2 ) .
  • Iterate the previous steps M times. Obtain the average value of the lower and upper bounds of κ , α , δ 1 , and δ 2 .

5. Simulation Study

We generate M = 1000 samples of size n = 25, 30, 40, 50, 60, 70, 80, 90, 100 from the ASKUG-LOMAX model via the initial parameter values κ ( 0 ) = 0.5 , α ( 0 ) = 0.9 , δ 1 ( 0 ) = 1.89 , and δ 2 ( 0 ) = 1.1 . Suppose that κ , α , δ 1 , and δ 2 are random variables that follow the prior PDFs Gamma ( κ ; 0.05 , 0.68 ) , Gamma ( α ; 0.6 , 0.9 ) , Gamma ( δ 1 ; 0.8 , 1 ) , and Gamma ( δ 2 ; 0.5 , 0.69 ) , respectively. In this simulation study, we empirically obtain the bias and expected risk (ER) of the MLEs and the Bayesian methods for different parameter combinations and each sample. The point estimations of the parameters are obtained using 200 burns MCMC methods. Two LINEX loss function are used, L E 1 when ρ = 0.3 and L E 2 when ρ = 0.7 . The respective biases and ERs are provided by
B i a s ϑ ^ = 1 1000 i = 1 1000 ϑ ^ i ϑ ,
and
E R ϑ ^ = 1 1000 i = 1 1000 ϑ ^ i ϑ 2 ,
Coverage probabilities (CPs) are calculated at the 95% and 90% HPD credible intervals. The simulation results are presented in Table 2, Table 3, Table 4 and Table 5 for the parameters κ , α , δ 1 , and δ 2 , respectively. Based on the generated data, it can be seen that
  • The MLEs seem to behave as expected, i.e., the MSE values and the estimated biases decrease as n increases. Moreover, the mean values of the estimates tend to the true values as n increases, showing the consistency property of the MLEs.
  • It is well known that the Bayesian estimation method provides better results in practice than the classical one, especially when the sample size is relatively small, which is exactly what the results show. The standard deviation of the MLE is greater than the Bayesian estimate for n 90 .
  • The interval width of the MLE for a given confidence level is greater than the Bayesian estimate in most cases.
  • A General Entropy Loss Function is a suitable alternative to the Modified LINEX loss function. The approximate Bayes estimate of the parameters based on the general entropy loss function provides better results in most cases.

Application of the ASKUG-Lomax Model

We evaluate the usefulness of the ASKUG-Lomax model by analyzing two examples of lifetime data.
Example 1. 
Industry lifetime data taken from [23]. The observations represent the reliability time of a coating machine: 1.00, 1.00, 5.00, 5.50, 12.50, 16.75, 17.75, 20.75, 22.50, 22.75, 25.00, 25.00, 27.25, 30.25, 43.75, 45.00, 48.00, 48.25, 97.50, 99.75, 136.75, 143.50, 207.75, 215.00, 225.50, 235.00, 283.50, 567.00, 970.50.
Example 2. 
Business data provided by Nigm and Hamdy [24] and Wong [25]: 1.01, 1.05, 1.08, 1.14, 1.28, 1.30, 1.33, 1.43, 1.59, 1.62.
After analyzing the data, the estimate results are
Example 3. 
The average per capita carbon dioxide emissions (in metric tons) in the Arab world, provided by [26]: 0.609268, 0.662618, 0.727117, 0.853116, 0.972381, 1.13867, 1.252, 1.31609, 1.45773, 1.76705, 1.79794, 1.99733, 2.11931, 2.19399, 2.2808, 2.40051, 2.5815, 2.64469, 2.71073, 2.75889, 2.80602, 2.86087, 2.87078, 2.91291, 2.9245, 2.95929, 2.97, 3.04479, 3.08222, 3.0918, 3.1313, 3.16119, 3.16314, 3.1633, 3.16669, 3.18374, 3.19605, 3.21329, 3.2388, 3.27108, 3.2779, 3.28251, 3.34723, 3.36173, 3.47424, 3.7043, 3.80165, 3.88314, 4.09354, 4.19327, 4.30874, 4.322, 4.43872, 4.49519, 4.51219, 4.52835, 4.57031, 4.60019, 4.61796.
The data are presented in Table 6, Table 7 and Table 8 for the parameters κ , α , δ 1 , and δ 2 , respectively. Table 9, Table 10 and Table 11 compare the ASKUG-Lomax distribution via several recognition criterion: the Akaike information criterion (AIC), Bayesian information criterion (BIC), Hannan–Quinn information criterion (HQIC), and consistent Akaike information Criterion (CAIC). The goodness-of-fit results of the ASKUG-Lomax model are compared with several other models in Table 1, including Arcsine exponentiated Lomax (ASEXG-Lomax), Arcsine Lomax (AS-Lomax), Exponentiated Weibull (EX-Weibull), and Weibull distribution.
The results in Table 9 indicate that the ASKUG-Lomax distribution provides better fits than the alternative models, and represents an adequate model for analyzing heavy-tailed industry claims data. In addition, the results in Table 10 indicate that the ASKUG-Lomax distribution provides better fits than other competing models, and is adequate for analyzing business data. The results in Table 11 again indicate that the ASKUG-Lomax distribution provides better fits than the other competing models, and is adequate for analyzing carbon dioxide emissions data.

6. Concluding Remarks

The use of the trigonometric arcsine function introduces a new family of heavy-tailed models, the Arcsine Kumaraswamy family of generalised X-distributions. The Arcsine Kumaraswamy-generalised X-distribution is very interesting, and provides a better fit for data with strong tails. A special submodel called the Arcsine Kumaraswamy-Lomax model is defined is this paper. We calculate the parameters of the Arcsine Kumaraswamy-generalised Lomax model with maximum likelihood, bootstrap, and Bayesian estimators. A simulation study and analysis of real industry data are provided to verify the performance of the Arcsine Kumaraswamy-Lomax model. The performance of the Bayesian estimators is better than that of the corresponding ML estimators. Based on our modeling of three real datasets, the results show that the Arcsine Kumaraswamy-generalised Lomax model provides a better fit than other competing models.

Author Contributions

Data curation, W.E.; Funding acquisition, Y.T.; Investigation, W.E.; Methodology, W.E.; Resources, W.E.; Supervision, W.E.; Writing—review & editing, Y.T. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by Researchers Supporting Project number (RSP2022R488), King Saud University, Riyadh, Saudi Arabia.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

All the datasets used in this paper are available from the corresponding author upon request.

Conflicts of Interest

The authors declare that they have no conflict of interest.

Appendix A

L δ 1 = n δ 1 + i = 1 n Log G x i ; φ + i = 1 n ( δ 2 1 ) Log G x i ; φ G x i ; φ δ 1 1 G x i ; φ δ 1 δ 2 2 1 1 G x i ; φ δ 1 δ 2 1 i = 1 n δ 2 Log G x i ; φ G x i ; φ δ 1 1 G x i ; φ δ 1 δ 2 1 1 G x i ; φ δ 1 δ 2 1 1 1 1 G x i ; φ δ 1 δ 2 2 ,
L δ 2 = n δ 2 + i = 1 n Log 1 G x i ; φ δ 1 1 G x i ; φ δ 1 δ 2 1 1 1 G x i ; φ δ 1 1 + δ 2 i = 1 n Log 1 G x i ; φ δ 1 1 G x i ; φ δ 1 δ 2 1 1 G x i ; φ δ 1 δ 2 1 1 1 G x i ; φ δ 1 δ 2 2 .
The ASKUG-Lomax log-likelihood function is
L = n Log 2 π κ α δ 1 δ 2 ( 1 + α ) i = 1 n Log 1 + κ x i + δ 1 1 i = 1 n Log 1 1 + κ x i α + δ 2 1 i = 1 n Log 1 1 1 + κ x i α δ 1 1 2 i = 1 n Log 1 1 1 1 1 + x i α β δ 1 δ 2 2 .
Maximize Equation (15) to obtain the MLE estimates, as follows:
L κ = n κ ( α + 1 ) i = 1 n x i 1 + κ x i + α δ 1 1 i = 1 n x i 1 + κ x i 1 α 1 1 + κ x i α α δ 1 δ 2 1 i = 1 n x i 1 + κ x i α 1 1 1 + κ x i α δ 1 1 1 1 1 + κ x i α δ 1 + i = 1 n α δ 1 δ 2 x i 1 1 1 + κ x i α δ 1 1 + δ 2 1 + 2 1 1 1 + κ x i α δ 1 δ 2 1 + κ x i α + 1 1 1 + κ x i α 1 δ 1 1 1 1 1 1 + κ x i α δ 1 δ 2 2 ,
L α = n α i = 1 n Log 1 + κ x i + δ 1 1 i = 1 n 1 + κ x i α Log 1 + κ x i 1 1 + κ x i α δ 1 δ 2 1 i = 1 n Log 1 + κ x i 1 + κ x i α 1 1 + κ x i α 1 + δ 1 1 1 1 + κ x i α δ 1 + i = 1 n δ 1 δ 2 Log 1 + κ x i 1 1 + κ x i α 1 + δ 1 1 + 2 1 1 1 + κ x i α δ 1 δ 2 1 + κ x i α 1 1 1 + κ x i α δ 1 δ 2 + 1 1 1 1 1 1 + κ x i α δ 1 δ 2 2 ,
L δ 1 = n δ 1 + i = 1 n Log 1 1 + κ x i α δ 2 1 i = 1 n Log 1 1 + κ x i α 1 1 + κ x i α δ 1 1 1 1 + κ x i α δ 1 ,
and
L δ 2 = n δ 2 + i = 1 n Log 1 1 1 + κ x i α δ 1 .

References

  1. Klugman, S.A.; Panjer, H.H.; Willmot, G.E. Loss Models: From Data to Decisions; John Wiley & Sons: Hoboken, NJ, USA, 2012; Volume 715. [Google Scholar]
  2. Lane, M.N. Pricing risk transfer transactions. ASTIN Bull. 2000, 30, 259–293. [Google Scholar] [CrossRef] [Green Version]
  3. Cooray, K.; Ananda, M. Modeling actuarial data with a composite lognormal-pareto model. Scand. Actuar. J. 2005, 2005, 321–334. [Google Scholar] [CrossRef]
  4. Ibragimov, R.; Prokhorov, A. Heavy tails and copulas: Topics in dependence modelling in economics and finance. Quant. Financ. 2019, 19, 13–14. [Google Scholar]
  5. Bernardi, M.; Maruotti, A.; Petrella, L. Skew mixture models for loss distributions: A bayesian approach. Insur. Math. Econ. 2012, 51, 617–623. [Google Scholar] [CrossRef] [Green Version]
  6. Adcock, C.; Eling, M.; Loperfido, N. Skewed distributions in finance and actuarial science: A review. Le Eur. J. Financ. 2015, 21, 1253–1281. [Google Scholar] [CrossRef]
  7. Bhati, D.; Ravi, S. On generalized log-moyal distribution: A new heavy tailed size distribution. Insur. Math. Econ. 2018, 79, 247–259. [Google Scholar] [CrossRef]
  8. Beirlant, J.; Matthys, G.; Dierckx, G. Heavy-tailed distributions and rating. ASTIN Bull. 2001, 31, 37–58. [Google Scholar] [CrossRef] [Green Version]
  9. Resnick, S.I. Discussion of the Danish data on large fire insurance losses. ASTIN Bull. 1997, 27, 139–151. [Google Scholar] [CrossRef] [Green Version]
  10. Dutta, K.; Perry, J. A tale of tails: An empirical analysis of loss distribution models for estimating operational risk capital. SSRN Electron. J. 2006, 1, 6–13. [Google Scholar] [CrossRef] [Green Version]
  11. Afify, A.Z.; Cordeiro, G.M.; Maed, M.E.; Alizadeh, M.; Al-Mofleh, H.; Nofal, Z.M. The generalized odd Lindley-G family: Properties and applications. An. Acad. Bras. Ciˆencias 2019, 91, 1–22. [Google Scholar] [CrossRef] [Green Version]
  12. Afify, A.Z.; Alizadeh, M. The odd Dagum family of distributions: Properties and applications. J. Appl. Probab. Stat. 2020, 15, 45–72. [Google Scholar]
  13. Nasir, A.; Yousof, H.M.; Jamal, F.; Korkmaz, M.Ç. The exponentiated Burr XII power series distribution: Properties and applications. Stats 2019, 2, 15–31. [Google Scholar] [CrossRef]
  14. Jamal, F.; Nasir, M. Some new members of the TX family of distributions. In Proceedings of the 17th International Conference on Statistical Sciences, Lahore, Pakistan, 21–23 January 2019; Volume 17. [Google Scholar]
  15. Al-Mofleh, H. On generating a new family of distributions using the tangent function. Pak. J. Stat. Oper. Res. 2018, 14, 471–499. [Google Scholar] [CrossRef] [Green Version]
  16. Afify, A.Z.; Cordeiro, G.M.; Butt, N.S.; Ortega, E.M.M.; Suzuki, A.K. A new lifetime model with variable shapes for the hazard rate. Braz. J. Probab. Stat. 2017, 31, 516–541. [Google Scholar] [CrossRef]
  17. Cordeiro, G.M.; Afify, A.Z.; Ortega, E.M.M.; Suzuki, A.K.; Mead, M.E. The odd Lomax generator of distributions: Properties, estimation and applications. J. Comput. Appl. Math. 2019, 347, 222–237. [Google Scholar] [CrossRef]
  18. Mudholkar, G.S.; Srivastava, D.K. Exponentiated Weibull family for analyzing bathtub failure-rate data. IEEE Trans. Reliab. 1993, 42, 299–302. [Google Scholar] [CrossRef]
  19. Cordeiro, G.M.; de Castro, M. A new family of generalized distributions. J. Stat. Comput. Simul. 2011, 81, 883–898. [Google Scholar] [CrossRef]
  20. Mead, M.E.; Afify, A.Z. On five-parameter Burr XII distribution: Properties and applications. S. Afr. Stat. J. 2017, 51, 67–80. [Google Scholar] [CrossRef]
  21. Mansour, M.M.; Aryal, G.; Afify, A.Z.; Ahmad, M. The Kumaraswamy exponentiated Fréchet distribution. Pak. J. Stat. 2018, 34, 177–193. [Google Scholar]
  22. Ahmad, Z.; Hamedani, G.G.; Butt, N.S. Recent developments in distribution theory: A brief survey and some new generalized classes of distributions. Pak. J. Stat. Oper. Res. 2019, 15, 87–110. [Google Scholar] [CrossRef] [Green Version]
  23. AL-Jamal, Z.Y. Exponentiated exponential distribution as a failure time distribution. Iraqi J. Stat. Sci. 2008, 8, 63–75. [Google Scholar] [CrossRef]
  24. Nigm, A.M.; Hamdy, H.I. Bayesian prediction bounds for the Pareto lifetime model. Commun. Stat. Theory Methods 1987, 16, 1761–1772. [Google Scholar] [CrossRef]
  25. Wong, A. Approximate studentization for Pareto distribution with application to censored data. Stat. Pap. 1998, 39, 189–201. [Google Scholar] [CrossRef]
  26. Available online: https://databank.worldbank.org/reports.aspx?source=world-development-indicators (accessed on 20 October 2022).
Table 1. New submodules via the ASKUG-X family.
Table 1. New submodules via the ASKUG-X family.
No.ModelDistribution FunctionGenerated ModelRange
1Beta 2 π a r c s i n 1 1 I x ( a , b ) δ 1 δ 2 ASKUG-Beta a < x < b , a , b , δ 1 , δ 2 > 0
2Burr 2 π a r c s i n 1 1 1 ( 1 + x a ) b δ 1 δ 2 ASKUG-Burr x 0 , a , b , δ 1 , δ 2 > 0
3Erlang 2 π a r c s i n 1 1 ( 1 / ( a 1 ) ! ) γ ( a , b x ) δ 1 δ 2 ASKUG-Erlang x 0 , a , b , δ 1 , δ 2 > 0
4Exponential 2 π a r c s i n 1 1 1 exp ( a x ) δ 1 δ 2 ASKUG-Exponential x 0 , a , δ 1 , δ 2 > 0
5Frechet 2 π a r c s i n 1 1 exp ( ( x / b ) a ) δ 1 δ 2 ASKUG-Frechet x 0 , a , b , δ 1 , δ 2 > 0
6Gamma 2 π a r c s i n 1 1 γ ( a , b x ) / G a m m a ( a ) δ 1 δ 2 ASKUG-Gamma x 0 , a , b , δ 1 , δ 2 > 0
7Gumbel 2 π a r c s i n 1 1 exp ( exp ( ( x a ) / b ) ) δ 1 δ 2 ASKUG-Gumbel x , a R , b , δ 1 , δ 2 > 0
8Half logistic 2 π a r c s i n 1 1 ( 1 exp ( x ) ) / ( 1 + exp ( x ) ) δ 1 δ 2 ASKUG-Half logistic x 0 , δ 1 , δ 2 > 0
9Kumaraswamy 2 π a r c s i n 1 1 1 ( 1 x a ) b δ 1 δ 2 ASKUG-Kumaraswamy 0 < x < 1 , a , b , δ 1 , δ 2 > 0
10Lindely 2 π a r c s i n 1 1 1 ( ( exp ( a x ) ( 1 + a + a x ) ) / ( 1 + a ) ) δ 1 δ 2 ASKUG-Lindely x 0 , a , δ 1 , δ 2 > 0
11Linear failure rate 2 π a r c s i n 1 1 1 exp ( a x b c x ) δ 1 δ 2 ASKUG-Linear failure rate x 0 , a , b , c , δ 1 , δ 2 > 0
12Log logistics 2 π a r c s i n 1 1 1 / ( 1 + ( x / b ) a ) δ 1 δ 2 ASKUG-Log logistics x 0 , a , b , δ 1 , δ 2 > 0
13Lomax 2 π a r c s i n 1 1 1 ( 1 + a x ) b δ 1 δ 2 ASKUG-Lomax x 0 , a , b , δ 1 , δ 2 > 0
14Normal 2 π a r c s i n 1 1 ϕ ( ( x a ) / b ) δ 1 δ 2 ASKUG-Normal x , a R , b , δ 1 , δ 2 > 0
15Pareto 2 π a r c s i n 1 1 1 ( x m / x ) a δ 1 δ 2 ASKUG-Pareto x m x < , a , x m , δ 1 , δ 2 > 0
16Power function 2 π a r c s i n 1 1 ( x / b ) a δ 1 δ 2 ASKUG-Power function 0 < x < b , a , b , δ 1 , δ 2 > 0
17Rayleigh 2 π a r c s i n 1 1 1 exp ( a x 2 ) δ 1 δ 2 ASKUG-Rayleigh x 0 , a , δ 1 , δ 2 > 0
18Topp Leone 2 π a r c s i n 1 1 x a ( 2 x a ) δ 1 δ 2 ASKUG-Topp Leone 0 < x < 1 , a , b , δ 1 , δ 2 > 0
19Uniform 2 π a r c s i n 1 1 x / a δ 1 δ 2 ASKUG-Uniform 0 < x < a , a , δ 1 , δ 2 > 0
20Weibull 2 π a r c s i n 1 1 1 exp ( a x b ) δ 1 δ 2 ASKUG-Weibull x 0 , a , b , δ 2 , δ 2 > 0
Table 2. Point and Interval estimation of the κ parameter when κ = 0.5 .
Table 2. Point and Interval estimation of the κ parameter when κ = 0.5 .
nPointIntrval
MLSELE1LE2GEMLBootstrapHPDSHPDLE1HPDLE2HPDGE
250.66610.64350.66050.60620.50180.2393 1.09280.06 1.8610.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.15210.12950.14650.0922−0.01210.85351.8010.3550.3550.33040.3561
0.04740.0260.03120.01680.00880.3068 1.02530.11 1.5980.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
0.71851.4880.3060.32740.29380.2894
300.64280.64350.66050.60620.5018−2.4957 3.78120.086 1.6890.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.12880.12950.14650.0922−0.01216.27691.6030.3550.3550.33040.3561
2.5640.0260.03120.01680.0088−1.9993 3.28480.137 1.5260.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
5.28411.3890.3060.32740.29380.2894
400.64760.64350.66050.60620.50180.0818 1.21340.12 1.9420.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.13360.12950.14650.0922−0.01211.13161.8220.3550.3550.33040.3561
0.08330.0260.03120.01680.00880.1713 1.12390.153 1.5540.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
0.95271.4010.3060.32740.29380.2894
500.61520.64350.66050.60620.50180.2347 0.99580.117 1.6960.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.10120.12950.14650.0922−0.01210.76111.5790.3550.3550.33040.3561
0.03770.0260.03120.01680.00880.2949 0.93560.162 1.4350.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
0.64071.2730.3060.32740.29380.2894
600.61520.64350.66050.60620.50180.1762 1.05420.117 1.6960.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.10120.12950.14650.0922−0.01210.8781.5790.3550.3550.33040.3561
0.05020.0260.03120.01680.00880.2456 0.98480.162 1.4350.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
0.73911.2730.3060.32740.29380.2894
700.64290.64350.66050.60620.50180.2584 1.02750.151 1.6990.467 0.8220.4798 0.83480.4375 0.76780.3198 0.6758
0.12890.12950.14650.0922−0.01210.76911.5480.3550.3550.33040.3561
0.03850.0260.03120.01680.00880.3192 0.96670.197 1.4920.492 0.7980.4958 0.82320.4663 0.76010.3605 0.6499
0.64751.2950.3060.32740.29380.2894
800.63560.60350.61160.58530.53520.3024 0.96880.161 1.7550.283 0.9990.2847 1.00670.278 0.97980.2518 0.9417
0.12160.08950.09760.07130.02130.66641.5940.7160.7220.70170.6899
0.02720.04070.04310.03580.0310.3551 0.930.211 1.480.336 0.9230.3387 0.93990.3313 0.88510.2889 0.8189
0.5611.2690.5870.60120.55370.53
900.65260.58690.59390.57080.52480.0865 1.21870.188 1.7430.304 1.0290.3067 1.04020.2988 0.99130.2691 0.9165
0.13860.07290.080.05680.01091.13221.5550.7250.73350.69250.6474
0.08340.0410.04310.03650.0330.176 1.12910.231 1.4750.341 0.9630.3438 0.9760.3346 0.93380.2922 0.8904
0.95311.2440.6220.63220.59920.5982
1000.65260.64490.65330.6260.57880.33 0.97520.188 1.7430.32 0.970.3234 0.98210.3127 0.94250.2904 0.911
0.13860.13090.13930.1120.06480.64531.5550.650.65870.62980.6206
0.02710.04750.05050.04120.03360.381 0.92420.231 1.4750.376 0.9330.3796 0.95370.3681 0.90290.3214 0.8833
0.54321.2440.5570.5740.53470.5619
Point Estimation: The first row represents the Estimate, the second row represents the Bias, and the third row represents the ER. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 3. Point and Interval estimation of the α parameter when α = 0.9 .
Table 3. Point and Interval estimation of the α parameter when α = 0.9 .
nPointIntrval
MLSELE1LE2GEMLBootstrapHPDSHPDLE1HPDLE2HPDGE
251.06571.01741.03420.97940.93150.8151 1.31640.369 1.9530.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.14560.09730.11410.05920.01140.50131.5840.6680.67620.64330.6343
0.01640.03320.03740.02590.02360.8548 1.27670.414 1.8260.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
0.4221.4120.4910.48970.46540.4746
301.06011.01741.03420.97940.93150.3908 1.72940.355 1.9710.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.13990.09730.11410.05920.01141.33871.6160.6680.67620.64330.6343
0.11660.03320.03740.02590.02360.4966 1.62360.401 1.8440.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
1.12691.4430.4910.48970.46540.4746
401.06031.01741.03420.97940.93150.8153 1.30540.389 1.8990.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.14020.09730.11410.05920.01140.49011.510.6680.67620.64330.6343
0.01560.03320.03740.02590.02360.854 1.26660.432 1.7890.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
0.41261.3570.4910.48970.46540.4746
501.09661.01741.03420.97940.93150.8639 1.32930.393 1.9510.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.17640.09730.11410.05920.01140.46541.5580.6680.67620.64330.6343
0.01410.03320.03740.02590.02360.9007 1.29250.433 1.80.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
0.39181.3670.4910.48970.46540.4746
601.09661.01741.03420.97940.93150.7381 1.45520.393 1.9510.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.17640.09730.11410.05920.01140.71711.5580.6680.67620.64330.6343
0.03350.03320.03740.02590.02360.7948 1.39850.433 1.80.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
0.60371.3670.4910.48970.46540.4746
701.0311.01741.03420.97940.93150.8339 1.2280.4 1.8510.664 1.3320.6703 1.34650.6513 1.29460.6241 1.2583
0.11080.09730.11410.05920.01140.39411.4510.6680.67620.64330.6343
0.01010.03320.03740.02590.02360.8651 1.19690.434 1.7470.789 1.280.8029 1.29260.7589 1.22420.7142 1.1888
0.33181.3130.4910.48970.46540.4746
801.07981.01931.02421.00740.99420.892 1.26750.395 1.870.584 1.5460.585 1.55350.579 1.5280.5677 1.5169
0.15960.09910.10410.08730.07410.37551.4750.9620.96860.94890.9492
0.00920.08290.08480.07860.07720.9217 1.23780.442 1.7630.67 1.5220.6717 1.53240.6654 1.49360.6554 1.4919
0.31611.3210.8520.86080.82820.8365
901.05161.00251.00580.99480.98560.8093 1.29390.395 1.8230.597 1.4950.5993 1.50190.5916 1.47860.5786 1.4707
0.13140.08240.08570.07460.06550.48461.4280.8980.90260.8870.8921
0.01530.06330.06430.06120.06090.8476 1.25560.435 1.7390.669 1.3640.6716 1.36520.6619 1.36090.649 1.3591
0.4081.3040.6950.69350.6990.7101
1001.05160.97370.97890.96170.94750.8846 1.21850.395 1.8230.594 1.5180.5945 1.52360.5914 1.50510.5859 1.4979
0.13140.05350.05870.04150.02730.33381.4280.9240.92910.91360.912
0.00730.07740.07890.07430.07410.911 1.19210.435 1.7390.621 1.4570.6223 1.45970.6188 1.45140.6131 1.4485
0.2811.3040.8360.83740.83260.8354
Point Estimation: The first row represents the Estimate, the second row represents the Bias, and the third row represents the ER. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 4. Point and Interval estimation of the δ 1 parameter when δ 1 = 1.89 .
Table 4. Point and Interval estimation of the δ 1 parameter when δ 1 = 1.89 .
nPointIntrval
MLSELE1LE2GEMLBootstrapHPDSHPDLE1HPDLE2HPDGE
252.11441.82191.85861.73951.71871.1013 3.12750.876 3.691.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.2245−0.068−0.0313−0.1504−0.17132.02622.8140.6070.59030.59780.6115
0.26720.02720.02330.04530.05381.2616 2.96730.992 3.5151.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
1.70572.5230.4880.47980.50010.5324
302.09151.82191.85861.73951.7187−3.0784 7.26130.885 3.6751.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.2015−0.068−0.0313−0.1504−0.171310.33972.790.6070.59030.59780.6115
6.95740.02720.02330.04530.0538−2.2607 6.44361.025 3.4751.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
8.70432.450.4880.47980.50010.5324
402.1061.82191.85861.73951.71871.4116 2.80040.968 3.6341.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.216−0.068−0.0313−0.1504−0.17131.38882.6660.6070.59030.59780.6115
0.12550.02720.02330.04530.05381.5214 2.69061.09 3.4641.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
1.16922.3740.4880.47980.50010.5324
502.0681.82191.85861.73951.71871.3319 2.80410.963 3.6241.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.1781−0.068−0.0313−0.1504−0.17131.47222.6610.6070.59030.59780.6115
0.1410.02720.02330.04530.05381.4484 2.68771.078 3.3851.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
1.23932.3070.4880.47980.50010.5324
602.0681.82191.85861.73951.71871.3888 2.74730.963 3.6241.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.1781−0.068−0.0313−0.1504−0.17131.35852.6610.6070.59030.59780.6115
0.12010.02720.02330.04530.05381.4962 2.63981.078 3.3851.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
1.14362.3070.4880.47980.50010.5324
702.09561.82191.85861.73951.71871.5005 2.69071.093 3.5451.498 2.1051.5513 2.14161.4204 2.01831.3967 2.0081
0.2056−0.068−0.0313−0.1504−0.17131.19022.4520.6070.59030.59780.6115
0.09220.02720.02330.04530.05381.5946 2.59661.166 3.2911.591 2.0791.6242 2.1041.4997 1.99981.4538 1.9863
1.00192.1250.4880.47980.50010.5324
802.10421.92461.94761.87141.86121.5096 2.69881.114 3.5691.349 2.3771.3577 2.3891.316 2.34571.2991 2.3471
0.21430.03460.0577−0.0185−0.02871.18922.4551.0281.03131.02971.048
0.0920.07970.08110.07980.08381.6037 2.60481.22 3.3641.394 2.3641.4065 2.37471.3686 2.32791.3563 2.3297
1.00112.1440.970.96830.95930.9734
902.11361.91461.93691.86351.8531.7443 2.48291.157 3.5481.38 2.4031.3962 2.42391.3143 2.37591.2948 2.3782
0.22360.02460.047−0.0264−0.03690.73852.3911.0231.02781.06171.0834
0.03550.07810.07840.08070.08561.8027 2.42451.248 3.3241.54 2.3861.5627 2.40481.4904 2.33831.4683 2.3402
0.62172.0760.8460.84210.84790.8719
1002.11361.97621.99151.94161.93641.5508 2.67641.157 3.5481.259 2.5371.2657 2.54251.2445 2.52381.2361 2.5257
0.22360.08630.10150.05160.04651.12562.3911.2781.27671.27931.2896
0.08250.11880.12090.1160.11911.6398 2.58741.248 3.3241.406 2.4911.4252 2.49861.3644 2.47421.3449 2.4765
0.94762.0761.0851.07341.10981.1316
Point Estimation: The first row represents the Estimate, the second row represents the Bias, and the third row represents the ER. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 5. Point and Interval estimation of the δ 2 parameter when δ 2 = 1.1 .
Table 5. Point and Interval estimation of the δ 2 parameter when δ 2 = 1.1 .
nPointIntrval
MLSELE1LE2GEMLBootstrapHPDSHPDLE1HPDLE2HPDGE
251.34471.09611.12141.03930.97310.8466 1.84280.431 2.9040.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.2108−0.0378−0.0125−0.0946−0.16080.99622.4730.6970.7250.62430.6022
0.06460.03010.03020.03410.05070.9254 1.7640.474 2.7270.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
0.83872.2530.5490.56320.51980.514
301.36361.09611.12141.03930.9731−0.2178 2.9450.444 2.910.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.2297−0.0378−0.0125−0.0946−0.16083.16272.4660.6970.7250.62430.6022
0.65090.03010.03020.03410.05070.0324 2.69480.486 2.7710.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
2.66252.2850.5490.56320.51980.514
401.31711.09611.12141.03930.97310.852 1.78220.46 2.850.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.1832−0.0378−0.0125−0.0946−0.16080.93022.390.6970.7250.62430.6022
0.05630.03010.03020.03410.05070.9256 1.70860.512 2.7130.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
0.78312.2010.5490.56320.51980.514
501.26881.09611.12141.03930.97310.8648 1.67270.488 2.8180.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.1348−0.0378−0.0125−0.0946−0.16080.80792.330.6970.7250.62430.6022
0.04250.03010.03020.03410.05070.9287 1.60880.532 2.6160.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
0.68012.0840.5490.56320.51980.514
601.26881.09611.12141.03930.97310.6134 1.92410.488 2.8180.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.1348−0.0378−0.0125−0.0946−0.16081.31072.330.6970.7250.62430.6022
0.11180.03010.03020.03410.05070.7171 1.82050.532 2.6160.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
1.10342.0840.5490.56320.51980.514
701.3141.09611.12141.03930.97310.9296 1.69850.499 2.6820.718 1.4150.7271 1.4520.7003 1.32460.6627 1.2649
0.1801−0.0378−0.0125−0.0946−0.16080.76892.1830.6970.7250.62430.6022
0.03850.03010.03020.03410.05070.9904 1.63770.541 2.550.831 1.380.8531 1.41630.7842 1.30410.7133 1.2273
0.64732.0090.5490.56320.51980.514
801.25161.10651.11571.08531.06510.9197 1.58350.494 2.8160.654 1.6540.6615 1.66470.6362 1.62850.5971 1.6193
0.1177−0.0274−0.0182−0.0486−0.06880.66382.32211.00320.99231.0222
0.02870.08230.08330.08050.08460.9722 1.5310.522 2.5270.705 1.6160.7087 1.62470.694 1.59420.6705 1.5839
0.55882.0050.9110.9160.90020.9134
901.29831.10691.1141.09051.07560.8266 1.770.503 2.7990.651 1.620.6527 1.63710.6428 1.59050.6164 1.5803
0.1644−0.027−0.0199−0.0434−0.05830.94342.2960.9690.98440.94770.9639
0.05790.07710.07840.07420.0760.9012 1.69540.546 2.590.713 1.5750.7136 1.60280.7103 1.5550.7055 1.5479
0.79422.0440.8620.88920.84460.8425
1001.29831.16271.17311.13881.11720.975 1.62160.503 2.7990.647 1.6670.653 1.67680.6344 1.64670.6087 1.6402
0.16440.02880.03920.0049−0.01680.64652.2961.021.02381.01231.0315
0.02720.09330.09550.08890.09171.0262 1.57040.546 2.590.687 1.6620.6942 1.67470.6751 1.62490.6519 1.6131
0.54432.0440.9750.98040.94970.9612
Point Estimation: The first row represents the Estimate, the second row represents the Bias, and the third row represents the ER. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 6. Point and Interval estimation of industry data.
Table 6. Point and Interval estimation of industry data.
ParameterPointIntrval
MLSELE1LE2GEMLHPDSHPDLE1HPDLE2HPDGE
κ 0.64810.61120.62670.57750.48280.001 2.64720.231 0.9210.2375 0.93630.2184 0.89130.1926 0.8158
2.64620.690.69880.67290.6232
0.001 2.3310.335 0.9040.3459 0.92610.3141 0.84880.2159 0.7785
2.330.5690.58020.53460.5627
α 0.13890.1370.13710.13680.13350.001 0.30550.115 0.1560.1147 0.15590.1144 0.15550.111 0.1519
0.30450.0410.04130.04110.0409
0.001 0.27910.118 0.1550.1178 0.15530.1175 0.15490.1146 0.1516
0.27810.0370.03750.03750.037
δ 1 3.062.83732.86742.77092.78630.2895 5.83052.213 3.1692.2227 3.20242.1959 3.11572.1974 3.1235
5.54090.9560.97970.91980.926
0.7277 5.39232.379 3.1592.3859 3.19232.3496 3.08882.3539 3.1063
4.66450.780.80640.73920.7524
δ 2 27.207827.057727.09826.96927.05050.001 86.480226.476 27.47226.4907 27.525726.4499 27.393926.4738 27.4622
86.47920.9961.0350.94410.9884
0.001 77.105526.579 27.43426.6082 27.478926.5259 27.337926.5742 27.4268
77.10450.8550.87070.8120.8526
Point Estimation: The row represents the Estimate point. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 7. Point and Interval estimation of business data.
Table 7. Point and Interval estimation of business data.
ParameterPointIntrval
MLSELE1LE2GEMLHPDSHPDLE1HPDLE2HPDGE
κ 0.16270.16270.16270.16260.16190.0001 0.55610.159 0.1670.1589 0.16710.1588 0.1670.1583 0.1663
0.5560.0080.00820.00810.008
0.0001 0.49390.159 0.1670.1594 0.1670.1593 0.16690.1587 0.1661
0.49380.0080.00760.00760.0074
α 11.058911.02211.059110.936411.00519.0918 13.02610.796 11.15610.8501 11.194210.7056 11.089510.7738 11.1408
3.93410.360.34410.38390.367
9.403 12.714810.824 11.15310.8665 11.183710.7253 11.087210.8027 11.139
3.31190.3290.31720.36190.3363
δ 1 25.707225.61325.653825.524925.605322.9367 28.477725.357 25.78825.4052 25.831325.2631 25.696525.3485 25.7804
5.54090.4310.42610.43340.4319
23.3749 28.039525.384 25.77725.4354 25.817725.2997 25.686425.3742 25.7704
4.66450.3930.38230.38670.3962
δ 2 36.486736.48936.530936.392136.483230.4393 42.534136.196 36.6736.2273 36.712236.1252 36.586536.1917 36.6652
12.09480.4740.48480.46120.4735
31.3958 41.577636.23 36.66236.255 36.699836.1715 36.577136.2266 36.6573
10.18180.4320.44480.40560.4307
Point Estimation: The row represents the Estimate point. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 8. Point and Interval estimation of carbon dioxide emissions data.
Table 8. Point and Interval estimation of carbon dioxide emissions data.
ParameterPointIntrval
MLSELE1LE2GEMLHPDSHPDLE1HPDLE2HPDGE
κ 10.510.525710.528610.518910.52438.7048 12.29529.732 11.1579.734 11.15769.7262 11.15719.7305 11.1574
3.59051.4251.42361.43091.4269
8.9887 12.01139.914 11.0939.9162 11.11559.908 11.08669.9127 11.0866
3.02261.1791.19931.17861.1739
α 2.92.28772.29382.27392.2752.7406 3.05941.972 2.6781.9726 2.67781.9717 2.67771.9717 2.6777
0.31880.7060.70520.7060.7061
2.7658 3.03422.016 2.6372.0163 2.65912.0141 2.62722.014 2.629
0.26840.6210.64280.61310.615
δ 1 1.882.31562.33052.27852.27591.2522 2.50781.533 2.8031.5448 2.80391.4496 2.8011.4051 2.8014
1.25561.271.25911.35141.3964
1.3515 2.40851.702 2.7781.7191 2.77771.6523 2.77731.6337 2.7773
1.0571.0761.05861.1251.1436
δ 2 1.130.28130.28180.28050.27491.06 1.20.192 0.3730.1919 0.37270.1919 0.37260.1914 0.3722
0.13990.1810.18080.18080.1808
1.0711 1.18890.216 0.3580.2158 0.35850.2151 0.35680.2118 0.3515
0.11780.1420.14270.14170.1397
Point Estimation: The row represents the Estimate point. Interval Estimation: 95% and 90%, respectively. The first and second rows show the HPD credible interval and the corresponding width of the parameter, respectively.
Table 9. Relative quality of ASKUG-Lomax on the industry lifetime data vs. competing models.
Table 9. Relative quality of ASKUG-Lomax on the industry lifetime data vs. competing models.
ModelAICBICHQCCAIC
ASKUG-Lomax254.692256.358260.161256.404
ASEXG-Lomax272.857271.897268.755271.573
AS-Lomax266.395265.934263.661265.539
EX-Weibull338.517339.477342.619339.802
Weibull370.397370.858373.131371.253
Table 10. Relative quality of ASKUG-Lomax on the business data vs. competing models.
Table 10. Relative quality of ASKUG-Lomax on the business data vs. competing models.
ModelAICBICHQCCAIC
ASKUG-Lomax19.68327.68320.89318.355
ASEXG-Lomax31.75035.75032.65830.754
AS-Lomax38.69540.41039.30138.032
EX-Weibull82.17886.17883.08581.182
Weibull63.21964.93363.82462.555
Table 11. Relative quality of ASKUG-Lomax on the carbon dioxide emissions data vs. competing models.
Table 11. Relative quality of ASKUG-Lomax on the carbon dioxide emissions data vs. competing models.
ModelAICBICHQCCAIC
ASKUG-Lomax244.007244.748252.317247.251
ASEXG-Lomax297.611298.047303.843300.044
AS-Lomax315.456315.671319.611317.078
EX-Weibull254.981255.417261.214257.414
Weibull303.874304.088308.029305.496
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Emam, W.; Tashkandy, Y. The Arcsine Kumaraswamy-Generalized Family: Bayesian and Classical Estimates and Application. Symmetry 2022, 14, 2311. https://doi.org/10.3390/sym14112311

AMA Style

Emam W, Tashkandy Y. The Arcsine Kumaraswamy-Generalized Family: Bayesian and Classical Estimates and Application. Symmetry. 2022; 14(11):2311. https://doi.org/10.3390/sym14112311

Chicago/Turabian Style

Emam, Walid, and Yusra Tashkandy. 2022. "The Arcsine Kumaraswamy-Generalized Family: Bayesian and Classical Estimates and Application" Symmetry 14, no. 11: 2311. https://doi.org/10.3390/sym14112311

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop