Dawoud-Kibria Estimator for Beta Regression Model: Simulation and Application

被引:19
作者
Abonazel, Mohamed R. [1 ]
Dawoud, Issam [2 ]
Awwad, Fuad A. [3 ]
Lukman, Adewale F. [4 ]
机构
[1] Cairo Univ, Fac Grad Studies Stat Res, Dept Appl Stat & Econometr, Giza, Egypt
[2] Al Aqsa Univ, Dept Math, Gaza City, Palestine
[3] King Saud Univ, Coll Business Adm, Dept Quantitat Anal, Riyadh, Saudi Arabia
[4] Univ Med Sci, Biostat & Epidemiol, Ondo City, Nigeria
关键词
beta Kibria-Lukman estimator; beta ozkale-Kaciranlar estimator; beta ridge estimator; maximum likelihood; mean square; RIDGE-REGRESSION; PERFORMANCE;
D O I
10.3389/fams.2022.775068
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
The linear regression model becomes unsuitable when the response variable is expressed as percentages, proportions, and rates. The beta regression (BR) model is more appropriate for the variable of this form. The BR model uses the conventional maximum likelihood estimator (BML), and this estimator may not be efficient when the regressors are linearly dependent. The beta ridge estimator was suggested as an alternative to BML in the literature. In this study, we developed the Dawoud-Kibria estimator to handle multicollinearity in the BR model. The properties of the new estimator are derived. We compared the performance of the estimator with the existing estimators theoretically using the mean squared error criterion. A Monte Carlo simulation and a real-life application were carried out to show the benefits of the proposed estimator. The theoretical comparison, simulation, and real-life application results revealed the superiority of the proposed estimator.
引用
收藏
页数:12
相关论文
共 37 条
[21]   On the performance of the poisson and the negative binomial ridge predictors [J].
Kaciranlar, Selahattin ;
Dawoud, Issam .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2018, 47 (06) :1751-1770
[22]   A New Ridge-Type Estimator for the Linear Regression Model: Simulations and Applications [J].
Kibria, B. M. Golam ;
Lukman, Adewale F. .
SCIENTIFICA, 2020, 2020
[23]   Performance of some new ridge regression estimators [J].
Kibria, BMG .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2003, 32 (02) :419-435
[24]   Multicollinearity and misleading statistical results [J].
Kim, Jong Hae .
KOREAN JOURNAL OF ANESTHESIOLOGY, 2019, 72 (06) :558-569
[25]   A New Ridge-Type Estimator for the Gamma Regression Model [J].
Lukman, Adewale F. ;
Dawoud, Issam ;
Kibria, B. M. Golam ;
Algamal, Zakariya Y. ;
Aladeitan, Benedicta .
SCIENTIFICA, 2021, 2021
[26]   Modified ridge-type for the Poisson regression model: simulation and application [J].
Lukman, Adewale F. ;
Aladeitan, Benedicta ;
Ayinde, Kayode ;
Abonazel, Mohamed R. .
JOURNAL OF APPLIED STATISTICS, 2022, 49 (08) :2124-2136
[27]   A new estimator for the multicollinear Poisson regression model: simulation and application [J].
Lukman, Adewale F. ;
Adewuyi, Emmanuel ;
Mansson, Kristofer ;
Kibria, B. M. Golam .
SCIENTIFIC REPORTS, 2021, 11 (01)
[28]   Modified ridge-type estimator for the gamma regression model [J].
Lukman, Adewale F. ;
Ayinde, Kayode ;
Kibria, B. M. Golam ;
Adewuyi, Emmanuel T. .
COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, 2022, 51 (09) :5009-5023
[29]   Modified ridge-type estimator to combat multicollinearity: Application to chemical data [J].
Lukman, Adewale F. ;
Ayinde, Kayode ;
Binuomote, Samuel ;
Clement, Onate A. .
JOURNAL OF CHEMOMETRICS, 2019, 33 (05)
[30]   Adjusted R2 - type measures for beta regression model [J].
Mahmood, Shaimaa Waleed ;
Seyala, Noor Nawzat ;
Algamal, Zakariya Yahya .
ELECTRONIC JOURNAL OF APPLIED STATISTICAL ANALYSIS, 2020, 13 (02) :350-357