JFSJournal of Function Spaces2314-88882314-8896Hindawi10.1155/2020/13435961343596Research ArticleType II Half Logistic Kumaraswamy Distribution with ApplicationsZeinEldinRamadan A.12https://orcid.org/0000-0002-0902-8080HaqMuhammad Ahsan ul34https://orcid.org/0000-0003-1824-7311HashmiSharqa45https://orcid.org/0000-0001-9954-0311ElsehetyMahmoud6ElgarhyM.7LeivaHugo1Deanship of Scientific ResearchKing Abdulaziz UniversitySaudi Arabiakau.edu.sa2Faculty of Graduate Studies for Statistical ResearchCairo UniversityEgyptcu.edu.eg3College of Statistical and Actuarial SciencesUniversity of the PunjabLahorePakistanpu.edu.pk4Quality Enhancement CellNational College of ArtsLahorePakistannca.edu.pk5Lahore College for Women University (LCWU)LahorePakistanlcwu.edu.pk6King Abdulaziz UniversitySaudi Arabiakau.edu.sa7Valley High Institute for Management Finance and Information SystemsObourQaliubia 11828Egypt2020182020202027032020050620201820202020Copyright © 2020 Ramadan A. ZeinEldin et al.This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

In this paper, a new distribution with a unit interval named type II half logistic Kumaraswamy (TIIHLKw) distribution is proposed. Its density and distribution functions are presented using alternate expressions. This distribution is obtained by adding an extra parameter in the existing model to rise its ability fitting complex data sets. Some important statistical properties of TIIHLKw distribution are derived. The estimation of the parameters is obtained by numerous well-recognized approaches and simulation study confirmed the efficiencies of estimates such obtained. We apply the related model to practical datasets, and it is concluded that the proposed model is the best by model selection criteria than other competitive models.

Deanship of Scientific Research (DSR), King Abdul Aziz University, JeddahDF-287-305-1441
1. Introduction

Over the last few years, inspired by the increasing demand of probability distributions in many fields, many generalized distributions have been studied. Most of them are proposed by the addition of more parameters to well-known probability distributions that exist in literature to make them flexible. For instance, Haq et al.  proposed and studied the generalized odd Burr III (GOBIII) family of distributions along with its important characterizations. Ahmed  proposed a new model derived by the transformation of the baseline model. Different shapes of failure function can be formed such as increasing and bathtub. Some mathematical properties are obtained. So many research works have been done for proposing more flexible generalized probability distributions such as Alzaatreh et al. , Haq et al. , Hashmi et al. , Elgarhy et al. , and ZeinEldin et al. . Alshenawy  suggested a new one-parameter distribution. Several statistical properties and characteristics of the proposed distribution are derived along with estimation under Type II censoring. The research is concluded on the basis of a simulation study and real data analysis.

The Kumaraswamy (Kw) distribution was introduced and studied by Kumaraswamy  with unit interval, denoted by 𝐾w(a, b), with cumulative distribution function (cdf) is (1)Gx=11xab,and its related probability density function (pdf) is (2)gx=abxa11xab1,x,a,b>0.

The Kw density has one of these shapes depending upon parameter values, unimodal (a,b>1), bathtub (a,b<1), increasing (a>1 and b1), decreasing (D) (a1 and b>1), or steady (both a and b equal to 1).

The behavior of Kw distribution is analogous to the beta distribution but simpler due to closed-form of both its pdf and cdf. Boundary behavior and the major special models are also the same in both Beta and Kw distributions. This distribution could be a good substitute in situations wherein reality the bounds are finite i.e., (0, 1).

The Kw is originally developed as a lifetime distribution. Many authors studied and developed the generalizations of Kw distribution such as exponentiated Kw distribution studied by Lemonte et al. , El-Sherpieny, and Ahmed  proposed Kw Kw distribution, transmuted Kw distribution studied by Khan et al. , Sharma and Chakrabarty  studied size biased Kw distribution, George and Thobias  introduced Marshall-Olkin Kw distribution, exponentiated generalized Kw distribution studied by Elgarhy et al. , type II Topp Leone inverted Kw by ZeinEldin et al. , type I half logistic inverted Kw by ZeinEldin et al. , truncated inverted Kw by Bantan et al. , and Ghosh  introduced bivariate and multivariate weighted Kw distributions.

Hassan et al.  proposed Type II half logistic-G (TIIHL-G). The TIIHL-G distribution is expressed by its cdf given by (3)Fx;λ,ζ=10logGx;ζ2λeλt1+eλt2dt=2Gx;ζλ1+Gx;ζλ,x>0,λ>0,where Gx;ζ is cdf of baseline model with parameter vector ζ and Fx;λ,ζ is cdf derived by the T-X generator proposed by Alzaatreh . The pdf of the TIIHL-G family is given as (4)fx;λ,ζ=2λgx;ζGx;ζλ11+Gx;ζλ2,x>0,λ>0.

This article is dedicated to both its mathematical and application features. A significant portion is kept for the estimation of the parameters through various methods including maximum likelihood estimation (MLE), least-square estimation (LSE), weighted least square estimation (WLSE), percentiles estimation (PCE), and Cramer-von Mises estimation (CVE).

The core purpose of this research is to suggest a simpler and more flexible model called Type II Half Logistic Kw (TIIHLKw) distribution. This article is organized in the following manner: Section 2 is dedicated for the proposition of type II half logistic-Kw (TIIHLKw) distribution. Section 3 deals with the leading statistical properties of this model. Important binomial expansions of density and distribution functions are presented which involve binomial expansions. In Section 4, an extensive study of five different methods of estimation is carried out, with all derivations and detailed discussions. A comprehensive simulation study is conducted to compute the biases and efficiency for parameters and compare the performances of five estimation approaches stated above in the next section. Section 6 is devoted to the real data application of TIIHLKwD to show the importance of TIIHLKw distribution. Lastly, the conclusion is given in Section 7.

2. The TIIHLKw Distribution

In this section, we examine the usefulness, and flexibility of a new associate of type II half-logistic-G family having (0, 1), using Kw distribution as the baseline. The pdf and cdf of Kw distribution (2 shape parameters: a,b>0) are given as follows (5)gx;a,b=abxa11xab1,0<x<1,(6)Gx;a,b=11xab,0<x<1.

The random variable (r.v.) X follows TIIHLKw distribution if its cdf is obtained by inserting (6) in (3) (7)Fx;λ,a,b=211xabλ1+11xabλ,a,b,λ>0,0<x<1.

The pdf of TIIHLKw distribution is as follows (8)fx;λ,a,b=2λabxa11xab111xabλ11+11xabλ2,a,b,λ>0,0<x<1.

The survival function of TIIHLKw distribution is (9)F¯x;λ,a,b=111xabλ1+11xabλ=21+11xabλ1.

The failure rate function of TIIHLKwD is (10)hx;a,b=2λabxa11xab111xabλ1111xab2λ.

The flexibility of TIIHLKw distribution can be illustrated in Figures 1, 2, and 3. The pdf plots for the TIIHLKw distribution are given in Figures 1 and 2, and plots of hrf are given in Figure 3.

Pdf plot of TIIHLKwD.

Plots of some pdfs of TIIHLKwD.

Plots of some hrf of TIIHLKwD.

3. Some Mathematical Properties

The properties of TIIHLKwD are derived here. After this, we consider a r.v. X follows the pdf (8) and cdf (7).

3.1. Quantile Function

The quantile function of X is denoted by Qu, defined as (11)Qu=G1u2u1/λ,u0,1,the inverse function of G. is G1., given in (6).

Also, (12)xu=11uu21/λ1/b1/a.

Simulated values from TIIHLKwD can be utilized in the simulation study. The r. v. u follows rectangular distribution on the interval 0 to 1, i.e., xu=Qu follows TIIHLKw distribution. In particular, the first three quartiles are obtained by putting u=1/4,1/2, and 3/4, respectively, in (11) and (12).

On differentiation, we can have the density of quantile function (13)qu=1abλu2u1+1/λ12u+uu221u2u1/λ1+1/b11u2u1/λ1/b1+1/a.

3.2. Alternate Representation

Here, we present the expansion of the pdf and cdf for TIIHLKw distribution for further mathematical manipulation.

The binomial theorem, for β>0 and z<1, can be expressed as (14)1zβ1=i=01iβ1izi,

Then, by applying (13) in (8), the pdf of TIIHLKwD becomes (15)fx;λ,a,b=2λabxa11+11xabλ2i=01iλ1i1xabi+11,for a,λ,b>0,0<x<1.

Another form of pdf (8) can be obtained by means of the following expansion (16)1+zβ=i=01iβ+i1izifor z<1 and β>0. By applying (16) in pdf (8), we get (17)fx=i=0ηixa11xab111xabλi+11,where ηi=2abλ1ii+1. Now considering (13), we have (18)11xabλi+11=j=01jλi+11j1xabj.

Inserting this expansion in (17), we have pdf of the following form (19)fx=i,j=0ηi,jxa11xabj+11,where ηi,j=1jλi+11jηi.

Another formula can be formed from pdf (19), which is given in (20) using an infinite linear combination (20)fx=i,j,k=0ηi,j1kbj+11kxak+11,(21)fx=i,j,k=0ηi,j,kxak+11,where ηi,j,k=1kbj+11kηi,j.

Proposition 1.

Let “h” be a positive integer. The expansion of cdf can be written in following form: (22)Fxh=l=0hm=0ηl,m1xabm,where ηl,m=2h1l+mh+l1lλl+hm.

Proof.

The cdf Fxh is obtained, for h an integer by using (16). (23)Fxh=2h11xabλh1+11xabλh,Fxh=l=0h2h1lh+l1l11xabλl+h.

Once more binomial expansion is applied to Fxh and it can be written as (24)Fxh=l=0hm=02h1l+mh+l1lλl+hm1xabm.

The required result is obtained by combing some expression together, completing the proof.

3.3. Probability Weighted Moments:

The probability-weighted moments (PWMs) are used to study some more characteristics of the probability distribution. Under the specified setting discussed above, PWMs are denoted by τr,s, can be defined as (25)τr,s=EXrFxs=xrFxsfxdx.

The PWMs of TIIHLKw are obtained by substituting (21) and (22) into (25), as follows (26)τr,s=i,j,k=0ηi,j,kl=0sm=0ηl,m01xak+1+r11xabmdx.

Then, (27)τr,s=i,j,k=0ηi,j,kl=0sm=0ηl,mΓ1+bmΓ1+k+r/aaΓ2+k+bm+r/a,τr,s=i,j,k=0ηi,j,kl=0sm=0ηl,mβ1+bm,1+k+ra,where Γs=0+tsetdt,s>0. Also, βa,b is beta function.

3.4. Moments

The moments have a vital role in the study of the distribution and real data applications. Now, we get the rth moment for the TIIHLKwD. Under the specified assumptions, the rth moment is obtained as (28)μr=xrfxdx=01i,j=0ηi,jxr+a11xabj+11dx,(29)μr=i,j=0ηi,jβr+a,bj+1;r=1,2,3,,where ηi,j=1jλi+11jηi.

The mean (μ1) and variance (var) of TIIHLKw distribution can be derived as (30)μ1=i,j=0ηi,jβ1+a,bj+1,varX=i,j=0ηi,jβ2+a,bj+1i,j=0ηi,jβ1+a,bj+12.

Also, the coefficients of skewness and kurtosis of TIIHLKw distribution are given by (31)Sk=μ3μ23/2,K=μ4μ22.

The summary measures, mean, variance, skewness (Sk) and kurtosis (K) values are presented in Table 1. The plots of mean and var for the TIIHLKwD are given in Figure 4 and graphs of skewness and kurtosis are presented in Figure 5 for different parameter ranges.

Mean, variance, skewness (Sk), and kurtosis (K) values a=3.

λbMeanVarSkK
12.00.565560.03908-0.097022.30905
3.00.505430.034300.048562.38086
4.00.464780.030530.131052.45360
5.00.434690.027580.184402.51360
6.00.411140.025220.221792.56192
7.00.391980.023300.249482.60112

1.52.00.645660.02830-0.267232.53251
3.00.579740.02601-0.088522.52495
4.00.534440.023730.011412.56666
5.00.500620.021760.075592.61164
6.00.473990.020110.120392.65173
7.00.452250.018710.153472.68603

2.02.00.695400.02170-0.353502.69440
3.00.626800.02080-0.150402.62680
4.00.579100.01930-0.037902.64750
5.00.543100.018000.034002.68340
6.00.514700.016800.084002.71910
7.00.491400.015700.120902.75120

2.52.00.729800.01740-0.403102.80610
3.00.660000.01720-0.181502.69350
4.00.610700.01640-0.059702.70010
5.00.573400.015400.017802.73070
6.00.543800.014400.071602.76440
7.00.519500.013600.111202.79570

3.02.00.815000.00830-0.483903.03600
3.00.745100.00940-0.212302.81430
4.00.693700.00950-0.066102.79630
5.00.653800.009300.026102.82280
6.00.621600.009000.089802.85860
7.00.594900.008600.136402.89410

Graphs of mean and variance of TIIHLKw distribution.

Graphs of skewness and kurtosis of TIIHLKw distribution.

We see a monotonic variation in these measures caused by variation in parameters a,b, and λ.

3.5. Moment-Generating Function

The moment-generating function of X, using moments about the origin (29), is obtained as (32)MXt=trr!μr=i,j,r=0trr!ηi,jβr+a,bj+1.

3.6. Incomplete Moments

The rth incomplete moment of TIIHLKwD can be obtained by using (19) is (33)Ert=0txrfxdx=0ti,j=0ηi,jxr+a11xabj+11dx,Ert=i,j=0ηi,jβtr+a,bj+1,r=1,2,3,,where βt.,. is incomplete beta function.

3.7. The Mean Deviation

Following are expressions used to get mean deviation about mean and median, respectively (34)Μ1X=2μFμ2TμandΜ2X=μ2TMd,where Md = Median of X and Tq=qxfxdx is the initial incomplete moment. Now (35)Tμ=0μxfxdx=i,j=0ηi,jβμ1+a,bj+1.TMd=0Mdxfxdx=i,j=0ηi,jβMd1+a,bj+1.

3.8. Bonferroni and Lorenz Curves

Bonferroni and Lorenz curves are important applications of the first incomplete moment. These curves are mostly used in different fields of life such as economics, reliability analysis, demographic studies, life testing, life insurance, and medical technology. The Lorenz and Bonferroni curves are obtained, respectively, as follows (36)LFx=1EX0xtftdt=i,j=0ηi,jβx1+a,bj+1i,j=0ηi,jβ1+a,bj+1,and, (37)BFx=LFxFx=1+11xabλi,j=0ηi,jβx1+a,bj+1/i,j=0ηi,jβ1+a,bj+1211xabλ.

3.9. Order Statistics

In statistical theory, order statistics is widely applied and practiced. Let X1,X2,,Xn be r.vs. with their corresponding cdfs Fx. Let X1:n,X2:n,.,Xn:n be the related ordered r. sample of size n, then the density of rth order statistic is given as (38)fr:nx=Kv=0nr1vnrvfxFxv+r1,where K=1/βr,nr+1 and β(.,.) is the beta function. The pdf of the rth order statistic of TIIHLKwD is obtained by putting (21) and (22) in (38), changing h with v+k1,(39)fr:nx=Kv=0nrk,m=0ηυxak+111xabm,where (40)ηυ=l=0v+r1i,j=01k+l+m2v+r1λi+11jbj+11kv+r+l2lλl+v+r1mnrvηi.

More, the sth moment of rth order statistics for TIIHLKwD is given by (41)EXr:ns=xsfr:nxdx.

By substituting (39) in (41), we have (42)EXr:ns=1βr,nr+1v=0nrk,m=0ηυ01xs+ak+111xabmdx.

Then, (43)EXr:ns=1Br,nr+1v=0nrk,m=0ηυβs+ak+1,bm+1.

3.10. Rényi Entropy

Renyi (1961) proposed and used this measure. It can be obtained by (44)IδX=11δlogfxδdx,δ>0andδ1.

Applying binomial expansion (16) in (8) then fxδ can be written as (45)fxδ=i=0ηixδa11xaδb111xabλi+δδ,

So, the Rényi entropy of TIIHLKw distribution is as follows (46)IδX=11δlogi=0ηi01xδa11xaδb111xabλi+δδdx.=11δlogi,j=0ηi,j01xδa11xabj+δδdx,where ηi,j=1jλi+δδjηi. (47)IδX=11δlogi,j=0ηi,jβδa1+1a,bj+b1δ+1.

3.11. Stress-Strength Reliability

This subsection deals with the stress-strength parameter of TIIHLKw distribution. Let X1 be the strength of a structure with a stress X2, and if X1 follows TIIHLKwλ1,a1,b1 and X2 follows TIIHLKwλ2,a2,b2, provided X1 and X2 are statistically independent r.vs.,

Proposition 2.

Under the assumption discussed above, we have (48)R=ηβa1m+1a2,lb2+1.

Proof.

The reliability is defined by (49)R=PX2<X1=0f1x;λ1,a1,b1F2x;λ2,a2,b2dx,

Then, we can write (50)R=2i,j,k=01kηi,j01xa111xa1b1j+1111xa2b2λ2k+1dx.=η01xa1m+111xa2lb2dx,where η=2i,j,k,l,m=01k+l+mλ2k+1+1lb1j+11mηi,j.which completes the proof.

4. Inference

This section is dedicated to estimation aspects of TIIHLKw distribution, assuming that population parameters (a,b,λ) are unknown and can be estimated using different methods of estimation including ML, LS, WLS, PC, and CV.

4.1. Maximum Likelihood Estimation

For a random sample of x1,x2,x3,,xn from the TIIHLKwλ,a,b distribution, the log-likelihood function for Φ=λ,a,b is (51)logLΦ=nlog2abλ+a1i=1nlogxi+λ1i=1nlog11xiab+b1i=1nlog1xia+i=1nlog1+11xiabλ2.

The members of UΦ=Uλ,Ua,Ub are given below (52)Uλ=nλ+i=1nLog11xiabi=1n2Log11xiab11xiabλ1+11xiabλ,(53)Ua=na+i=1nLogxi+λ1i=1nbLogxixia1xia1+b11xiabb1i=1nLogxixia1xiai=1n2bλLogxixia1xia1+b11xiab1+λ1+11xiabλ,(54)Ub=nb+i=1nLog1xiaλ1i=1nLog1xia1xiab11xiab+i=1n2λLog1xia1xiab11xiab1+λ1+11xiabλ.equating (52), (53), and (54) to zero and solve them simultaneously give the ML estimates (MLEs) Φ^=λ,a,bT of Φ=λ,a,bT. The iterative algorithm is used to obtain the numerical solution of these nonlinear equations such as the Newton-Raphson method.

4.2. Ordinary and Weighted LS Estimation (LSEs and WLSEs):

The LSEs of λ,a, and b can be obtained by minimizing the sum of squares of errors with respect to parameters. Suppose X1,X2,,Xn is a random sample of size n from TIIHLKwD and suppose X1,X2,,Xn be the related ordered sample.

The sum is independent of the unknown parameters. (55)i=1nFixin+12,

Equivalently, (56)i=1n211xiabλ1+11xiabλin+12,

The following function is minimized with respect to λ,a and b to get WLS estimators of model parameters. (57)WLSλ,a,b=i=1nn+12n+2ini+1211xiabλ1+11xiabλin+12.

i.e., WLSλ,a,b/λ=0,WLSλ,a,b/a=0, and WLSλ,a,b/b=0.

4.3. PC Estimators (PCEs)

Under the specification defined above including order statistics having relationship X1<X2<<Xn. In the PC method of estimation, the estimators of λ,a, and b are derived by minimizing the following expression (58)i=1nlnin+1ln211xiabλ1+11xiabλ2.with respect to λ,a, and b.

4.4. The Cramer-von Mises Minimum Distance Estimation

The CVE is another estimation method based on minimum distance. The CV estimators are obtained by minimizing, with respect to λ,a, and b. (59)CV=112n+i=1n211xiabλ1+11xiabλ2i12n2.

CV minimum distance estimators provide empirical evidence that the bias of this estimator is smaller than the other minimum distance estimators.

5. Simulation Study

In this section, we present a Monte Carlo (MC) simulation study in order to illustrate the behavior of different estimates. We consider four random sample sizes: n=50, 100, 200, and 500 from the TIIHLKw distribution and the samples are drawn 10,000 times. Four specifications of the parameters are used in this simulation study, given as, set1=λ=2,b=0.5,a=2, set2=λ=3,b=2,a=2, set3=λ=2,b=3,a=2, set4=λ=2,b=2,a=1. For sample generated, the MLE, LSE, WLSE, CVE, PCE, and MPSE of estimators are computed numerically. Then, the estimates of all methods and their mean square errors (MSEs) are documented in Tables 14.

Estimates of parameters and their corresponding MSEs are calculated using the same sample under different methods of estimation for four sets of parameters mentioned above for small sample size (n=50) to sufficient large sample size, that is, n=500.

The entries of Table 2 to Table 5 show that estimates are reliable and consistent. MSEs reduce as sample size (n) increases under each method of estimation. The summary of these four tables is presented in Table 5.

Estimates and MSEs of parameters of TIIHLKw Distribution for the Set1 (a=2,λ=2,b=2).

nMLEsLSEsWLSEsCVEsPCEs
EstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSE
502.2850.8211.7330.2631.8330.4421.7510.2931.7270.317
1.970.4672.5660.9272.5861.3272.7761.3252.7471.31
2.0920.0771.9610.2582.0350.232.0990.2682.0780.372

1002.2260.3931.7410.2391.7550.3921.7030.2351.6160.304
1.9250.3292.5540.842.6511.2382.7131.1282.8101.221
2.0560.0311.9610.1231.9520.0932.0140.0991.9290.078

2001.8380.1871.7590.201.9750.3271.7100.2171.7000.236
2.2870.2472.5660.832.2950.8692.6110.9172.6340.905
1.950.0181.9960.0652.0010.081.9570.061.9670.069

5001.9010.0791.7090.2051.8720.261.7640.1731.7060.207
2.1750.1292.5720.8042.4170.8532.5030.7122.5640.783
1.9733.3771.930.0321.9910.0311.9630.0291.9210.029

10^-3.

Estimates and MSEs of parameters of TIIHLKwD for the Set2 (a=2,λ=3,b=2).

nMLEsLSEsWLSEsCVEsPCEs
EstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEs
502.2090.4171.9410.2911.950.3692.0690.3652.120.327
2.8330.4923.4611.3653.4991.8713.4541.493.191.119
2.0240.0442.020.1642.0330.252.170.2892.1710.264

1002.2580.3771.9730.2181.9780.3291.9760.2451.9740.24
2.6890.4493.3351.123.5041.7523.351.1273.31.01
2.020.0222.0340.0952.0360.1282.0150.0932.0050.113

2002.2070.1321.9860.1981.9380.3291.9460.2122.0310.218
2.7280.23.2851.0063.5591.7083.4911.2133.2770.99
2.0527.5392.0020.0462.0170.0552.0530.0622.0630.074

5002.140.031.950.1821.9610.3231.9790.1891.9580.186
2.780.073.3170.9253.4541.7073.2510.8963.2940.902
2.0393.8261.9850.0221.9910.0281.9850.0291.9850.025

Estimates and MSEs of parameters of TIIHLKwD for the Set3 (a=2,λ=2,b=3).

nMLEsLSEsWLSEsCVEsPCEs
EstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEs
500.2621.9091.7460.2851.9420.6891.8280.3731.7280.309
0.852.3762.6351.262.5691.5382.7021.4612.8441.888
0.1112.9242.9310.8633.0630.9413.1140.643.0551.001

1000.1972.1311.6710.2611.9270.4891.6860.261.7040.215
0.2311.9832.7171.112.5541.442.7311.1952.7651.32
0.0943.1342.7820.2583.0030.3172.8660.2862.8390.221

2000.1192.0071.7290.1991.9450.4831.7270.2111.8230.162
0.162.0632.5980.9622.4781.1672.6430.9682.5090.904
0.0462.9682.8080.1563.0340.272.8860.1942.9190.127

5000.0772.1451.7870.1582.0030.4681.7640.1661.7890.152
0.0991.8762.4640.6372.3910.9832.4970.7272.5310.793
0.0343.112.8630.1113.0720.2542.8360.112.9280.085

10^-3.

Estimates and MSEs of parameters of TIIHLKw Distribution for the Set4 (a=1,λ=2,b=2).

NMLEsLSEsWLSEsCVEsPCEs
EstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEsEstimatesMSEs
501.2510.21.3210.6141.3530.8171.3310.6451.4711.469
1.8850.4081.8220.741.8831.1031.8380.7921.8650.874
2.1040.0992.2460.5712.2360.7422.3791.0462.51.88

1001.2320.1571.1820.1951.2290.51.2260.3331.3410.501
1.6740.2391.8280.4631.970.7931.9540.7011.750.751
2.1310.0692.1450.2942.1160.1812.2470.5062.3070.403

2001.1780.0471.1640.111.1560.2291.230.1321.1810.1
1.660.1571.790.341.9620.4861.7130.3151.810.319
2.090.0112.0620.0952.1170.1192.1690.1232.1650.125

5001.1270.0221.1520.0771.1940.131.1410.0661.1120.058
1.7530.0891.7460.2221.7460.2611.7810.2011.8490.163
2.0685.9232.0920.0382.0980.0562.0660.0352.0650.034

10^-3.

Ranks of all the methods of estimation for different parametric specifications.

ParametersNMLLSWLSPCCV
a=2λ=2b=250214.534.5
10013.5523.5
20012534
50014523

a=2λ=3b=250214.54.53
1001.51.5543
2001253.53.5
50012543

a=2λ=2b=35012534
10013542
20013542
50013542

a=1λ=2b=25012435
100123.53.55
2001253.53.5
50014532

Ranks18.53876.55453

Overall rank12543

MLEs and goodness-of-fit measures for the first data set.

Modela^b^λ^LAICBICAW
TIIHLKw8.4511828229.60.20922957.6664-109.333-104.5970.3362290.014361
Kw2.0774033.137456.0687-108.137-104.3130.6884070.105252
Tkw2.0774016.56870.9999956.0687-106.137-100.4010.6884070.105252
SBKw1.4471619.966955.2067-106.413-102.5890.8380170.127612
Beta2.6825713.865854.6067-105.213-97.47720.9268500.155443

MLEs and goodness-of-fit measures for the second data set.

Modela^b^λ^LAICBICAW
TIIHLKw0.320085.84284222.76658.5465-111.093-105.4790.175310.025823
Kw2.7187444.660452.4915-100.983-97.24071.310570.208116
TKw2.9854243.92050.65154853.7972-101.594-95.98081.020320.155492
SBKw2.1950932.902453.6088-103.218-99.47521.104200.174441
Beta5.9417721.205755.6002-107.200-99.58680.789920.131428

Using the entries of Table 6 for different parametric combinations, we can conclude that the MLE method outperforms than all other estimation methods (with an overall score of 18.5). Therefore, depending on the simulation study, the MLE method performs best for TIIHLKwD.

6. Applications

The TIIHLKw distribution aims at providing an alternative distribution to fit data on the unit interval to other distributions available in the literature. Here, we used the following probability distributions as competitor models;

The Kumaraswamy (Kw) distribution (60)fx;a,b=abxa11xab1,0<x<1.

Transmuted Kumaraswamy (TKw) distribution (61)fTKwx;a,b=abxa11xab11λ+2λ1xab.

Size biased Kumaraswamy (SBKw) distribution (62)fSBKwx;a,b=axa1xab1Beta1+1/a,b.

Beta distribution (63)fBetax;a,b=1Betaa,bxa11xb1.

We use the following accuracy measures for model comparison: Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), log-likelihood, Anderson-Darling (A), and Cramer–von Mises (W). R-Language is used for numerical computations.

1st data set. The data set was taken from Dasgupta , and considered n=50 on burr (in millimeters), with hole diameter and a sheet thickness of 12 mm and 3.15 mm, respectively.

2nd data set. This data set was taken from , and considered n=48 measurements on petroleum rock samples from a petroleum reservoir.

The total test time (TTT) for both datasets are presented in Figure 6. We can observe that the shape of TTT plots is concave for both datasets, which demonstrates an increasing failure rate.

Total test time (TTT) for both data sets.

The MLEs for TIIHLKw distribution along with some adequacy measures are presented in Tables 7 and 8.

Hence, it is concluded that the new model provides the better fit. Figures 7 and 8 show the estimated densities, cdfs, estimated survival functions, and PP plots for the considered distributions of both data sets, respectively. We note that the proposed model is more appropriated to fit the data than the other competing models.

The fitted pdf, cdf, survival, and PP plots of the TIIHLKw distribution for the first data.

The fitted pdf, cdf, survival, and PP plots of the TIIHLKw distribution for second data.

7. Conclusion

In this article, a new Type-II Half Logistic Kumaraswamy distribution is proposed. Some characteristics of the TIIHLKw distribution including linear combination expressions for the density function, probability weighted moments, moments, incomplete moments, quantile function, mean deviation about mean and about median, Bonferroni and Lorenz curves, order statistics, stress-strength reliability, and Rényi entropy are derived. The ML method is used to estimate model parameters. An extensive simulation study is conducted to compare several well-known estimation methods, including the method of maximum likelihood estimation, methods of least squares and weighted least squares estimation, and method of Cramer-von Mises minimum distance estimation. The simulation study showed the reliability and efficiency of the estimates. Finally, by considering the method of maximum likelihood estimation, the new model is fitted to two practical data sets. The applications on real data sets validated the significance of the new distribution.

AppendixA.1. Data I

0.04, 0.02, 0.06, 0.12, 0.14, 0.08, 0.22, 0.12, 0.08, 0.26, 0.24, 0.04, 0.14, 0.16, 0.08, 0.26, 0.32, 0.28, 0.14, 0.16, 0.24, 0.22, 0.12, 0.18, 0.24, 0.32, 0.16, 0.14, 0.08, 0.16, 0.24, 0.16, 0.32, 0.18, 0.24, 0.22, 0.16, 0.12, 0.24, 0.06, 0.02, 0.18, 0.22, 0.14, 0.06, 0.04, 0.14, 0.26, 0.18, 0.16.

A.2. Data II

0.0903296, 0.2036540, 0.2043140, 0.2808870, 0.1976530, 0.3286410, 0.1486220, 0.1623940, 0.2627270, 0.1794550, 0.3266350, 0.2300810, 0.1833120, 0.1509440, 0.2000710, 0.1918020, 0.1541920, 0.4641250, 0.1170630, 0.1481410, 0.1448100, 0.1330830, 0.2760160, 0.4204770, 0.1224170, 0.2285950, 0.1138520, 0.2252140, 0.1769690, 0.2007440, 0.1670450, 0.2316230, 0.2910290, 0.3412730, 0.4387120, 0.2626510, 0.1896510, 0.1725670, 0.2400770, 0.3116460, 0.1635860, 0.1824530, 0.1641270, 0.1534810, 0.1618650, 0.2760160, 0.2538320, 0.2004470.

Data Availability

Data is present in appendix.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was funded by the Deanship of Scientific Research (DSR), King Abdul Aziz University, Jeddah, under grant No. (DF-287-305-1441). The authors gratefully acknowledge the DSR technical and financial support.

HaqM. A. u.ElgarhyM.HashmiS.The generalized odd Burr III family of distributions: properties, applications and characterizationsJournal of Taibah University for Science201913196197110.1080/16583655.2019.1666785Abd El-BarA. M. T.RagabI. E.On weighted exponential-Gompertz distribution: properties and applicationJournal of Taibah University for Science201913161662710.1080/16583655.2019.1600277AlzaatrehA.LeeC.FamoyeF.A new method for generating families of continuous distributionsMetron2013711637910.1007/s40300-013-0007-y2-s2.0-84882981547HaqM. A. u.ElgarhyM.HashmiS.OzelG.AinQ. u.Transmuted Weibull power function distribution: its properties and applicationsJournal of Data Science201839741810.6339/JDS.201804_16(2).0009HashmiS.HaqM. A. u.UsmanR. M.A generalized exponential distribution with increasing, decreasing and constant shape hazard curvesElectronic Journal of Applied Statistical Analysis2019121223244ElgarhyM.HaqM. A. u.AinQ. u.Exponentiated generalized Kumaraswamy distribution with applicationsAnnals of Data Science20185227329210.1007/s40745-017-0128-xZeinEldinR. A.Ahsan Ul HaqM.HashmiS.ElsehetyM.Alpha power transformed inverse Lomax distribution with different methods of estimation and applicationsComplexity2020202015186081310.1155/2020/1860813AlshenawyR.A new one parameter distribution: properties and estimation with applications to complete and type II censored dataJournal of Taibah University for Science2020141111810.1080/16583655.2019.1698276KumaraswamyP.A generalized probability density function for double-bounded random processesJournal of Hydrology1980461-2798810.1016/0022-1694(80)90036-02-s2.0-0018911706LemonteA. J.SouzaaW. B.CordeiroG. M.The exponentiated Kumaraswamy distribution and its log-transformBrazilian Journal of Probability and Statistics2013271315310.1214/11-BJPS1492-s2.0-84869826223El-SherpienyE.-S.AhmedM. A.On the Kumaraswamy Kumaraswamy distributionInternational Journal of Basic and Applied Sciences20143410.14419/ijbas.v3i4.3182KhanM. S.KingR.HudsonI. L.Transmuted Kumaraswamy distributionStatistics in Transition New Series201617218321010.21307/stattrans-2016-013SharmaD.ChakrabartyT. K.On size biased Kumaraswamy distributionStatistics, Optimization & Information Computing20164310.19139/soic.v4i3.2172-s2.0-84991102847GeorgeR.ThobiasS.Marshall-Olkin Kumaraswamy distributionInternational Mathematical Forum2017122476910.12988/imf.2017.611151ZeinEldinR. A.JamalF.ChesneauC.ElgarhyM.Type II Topp-Leone inverted Kumaraswamy distribution with statistical inference and applicationsSymmetry20191112145910.3390/sym11121459ZeinEldinR. A.ChesneauC.JamalF.ElgarhyM.Statistical properties and different methods of estimation for type I half logistic inverted Kumaraswamy distributionMathematics2019710100210.3390/math7101002BantanR. A.JamalF.ChesneauC.ElgarhyM.Truncated inverted Kumaraswamy generated family of distributions with applicationsEntropy20192111108910.3390/e21111089GhoshI.Bivariate and multivariate weighted Kumaraswamy distributions: theory and applicationsJournal of Statistical Theory and Applications201918319810.2991/jsta.d.190619.001HassanA. S.ElgarhyM.ShakilM.Type II half logistic family of distributions with applicationsPakistan Journal of Statistics and Operation Research201713224526410.18187/pjsor.v13i2.1560DasguptaR.On the distribution of burr with applicationsSankhya B201173111910.1007/s13571-011-0015-y2-s2.0-80755126809CordeiroG. M.dos Santos BritoR.The beta power distributionBrazilian Journal of Probability and Statistics20122618811210.1214/10-BJPS1242-s2.0-82755162719