International Journal of Nonlinear Analysis and Applications

The effects of outlier on some Bayesian survival estimators for Burr-X distribution with a Covid-19 as a case study

Document Type : Research Paper

Author

Middle Technical University, Technical College of Management --Baghdad, Information Technology Department, Iraq.

Abstract

Survival functions estimators can be affected by outlier, and thus these estimations move away from their real values, especially with the increasing in the outlier ratios within the sample of the random variable. The research included a comparison of a number of Bayesian methods for the estimations of survival functions of burr- X distribution with the percentages of different outliers within the sample. Simulation results showed the effect of the estimation methods by sample size and the percentage of outliers, and the real values of the parameters distribution.
    
    Mean square error was adopted as a measure to compare the estimation methods with a number of simulation experiments. The research also included a case study of Covid-19 for practical application. Other estimation methods can be taken (maximum likelihood estimation method, moment method, and shrinkage method) to note the possibility of being affected by outlier values

Keywords

[1] M. T. Ahmed, M. A. Khaleel, P. E. Oguntunde and M. K. Abdal-Hammed, editors, A New Version of the Exponentiated Burr X distribution, J. Phys.: Conf. Ser., 2021, IOP Publishing.
[2] F. G. Akg¨ul and B. S¸eno˘glu, Inferences for stress–strength reliability of Burr Type X distributions based on ranked set sampling, Commun. Stat.- Simul. Comput., (2020) 1-17.
[3] S. Bals, S. Van Aert, G. Van Tendeloo and D. Avila-Brande, Statistical estimation of atomic positions from exit wave reconstruction with a precision in the picometer range, Phys. Rev. Lett., 96(9)(2006) 096106.
[4] M. Eliwa, Z. Alhussain, E. Ahmed, M. Salah, H. Ahmed and M. El-Morshedy, Bivariate Gompertz generator of distributions: statistical properties and estimation with application to model football data, J. National Sci. Found. Sri Lanka., 48(2)(2020).
[5] J. Fan and W. Zhang, Statistical estimation in varying coefficient models, The Ann. Stat., 27(5)(1999) 1491-518.
[6] A. S. Hassan, A. M. Abd-Elfattah, M. M. Hassan, Bayesian analysis for mixture of Burr XII and Burr X distributions, Far East J. Math. Sci. , 103(6) (2018) 1031-1041.
[7] N. A. Ibrahim, M. A. Khaleel, F. Merovci, A. Kilicman and M. Shitan, Weibull Burr X distribution properties and application, Pak. J. Stat.,33(5)(2017) 315-336.
[8] T. Koopmans, Statistical estimation of simultaneous economic relations, J. Am. Stat. Assoc., 40(232)(1945) 448-466.
[9] B. Legried, E. K. Molloy, T. Warnow and S. Roch, Polynomial-time statistical estimation of species trees under gene duplication and loss. J. Comput. Biol., 28(5)(2021) 452-468.
[10] A. K. Mahto, Y. M. Tripathi and S-J. Wu, Statistical inference based on progressively type-II censored data from the Burr X distribution under progressive-stress accelerated life test, J. Stat. Comput. Simul., 91(2)(2021) 368-382.
[11] I. A. Norman and A. H. A. Abd Al-Ameer, Comparison of Classical and Bayesian methods to Estimate the shape parameter and Reliability function in Burr type X or two-parameter of exponential Rayleigh distribution under different Loss function, J. Administration Econ., 42 (119) (2019) 42-58.
[12] A. Rabie and J. Li, E-Bayesian estimation for Burr-X distribution based on generalized type-I hybrid censoring scheme, Am. J. Math. Management Sci., 39(1)(2020) 41-55.
[13] M. Ravikumar and R. Kantam, Two-step estimation in burr type X distribution, Int. J. Stat. Appl. Math., 1(4)(2016) 37-41.
[14] Z. Yang, N. Goldman and A. Friday, Maximum likelihood trees from DNA sequences: a peculiar statistical estimation problem, Syst. Biol., 44(3)(1995) 384-99.
[15] G. Yari and Z. Tondpour, The new Burr distribution and its application, Math. Sci., 11(1)(2017) 47-54.
[16] S. Zhou, Y. Chen and J. Shi, Statistical estimation and testing for variation root-cause identification of multistage manufacturing processes. IEEE Trans. Autom. Sci. Eng., 1(1)(2004) 73-83.
International Journal of Nonlinear Analysis and Applications
Volume 13, Issue 1
March 2022
Pages 2971-2983
  • Receive Date: 12 June 2021
  • Revise Date: 04 October 2021
  • Accept Date: 22 November 2021