The half-logistic (Type-I)-skew-t model : Comparison of different estimation approaches using Monte Carlo simulations

Adubisi, O. D.; Abdulkadir, A.; Chiroma, H.

doi:10.47974/JSMS-1226

Open Access Research Article

The half-logistic (Type-I)-skew-t model : Comparison of different estimation approaches using Monte Carlo simulations

^*O. D. AdubisiCorresponding authoradubisiobinna@fuwukari.edu.ngAffiliation 1Department of Mathematics and StatisticsFederal University WukariTaraba State, 670101, NigeriaAffiliation 2Department of Mathematical SciencesAbubakar Tafawa Balewa UniversityBauchi State, 740272, NigeriaView full profile → , A. AbdulkadirDepartment of Mathematical SciencesAbubakar Tafawa Balewa UniversityBauchi State, 740272, NigeriaView full profile → , H. ChiromaDepartment of Mathematical SciencesAbubakar Tafawa Balewa UniversityBauchi State, 740272, NigeriaView full profile →

* Corresponding author · click or hover a name for details

pp. 1433–1451Vol. 27Issue 7October 2024DOI: 10.47974/JSMS-1226XML

Received:: 11 Nov 2020
Published Online:: 30 Nov 2024
Article type:: Research Article
Language:: EN
Article no.:: JSMS-1226
Pages:: 1433–1451

Abstract

In the paper, we looked at the problem of employing several non-Bayesian estimation approaches to estimate the parameters of the Type-I half-logistic skew-t (TIHLST) model. The maximum likelihood, least squares, weighted least squares, Anderson-Darling, maximum product of spacing, and Cramer-von Mises are the six methods for TIHLST parameter estimation that we examined. The Monte Carlo simulations were used to assess the TIHLST parameter estimations’ finite sample performance. The partial and overall ranks of the estimating techniques for every combination of parameters were used to determine the ordering performance of the six criteria. Based on the overall ranks, the results indicate that the criteria performance is ranked from best to worst using the following estimators: maximum likelihood, weighted least squares, maximum product of spacing, least squares, and Cramer-von Mises for all parameter combinations. Based on an overall rank of 34, we infer that the maximum-likelihood criterion fared better than all other criteria. For the TIHLST model, we thereby affirm the supremacy of the maximum likelihood criterion with an overall rank of 34 and the Anderson-Darling criterion with an overall rank of 57.5.

Keywords

Simulation Maximum likelihood Cramer-von mises Skew-t distribution Type-I half-logistic-G

Subject Classifications

(2010) 60E0562F1062F1262E10

References

[1] H. G. Dikko and S. Agboola, “Exponentiated generalized skew-t distribution,” Journal of the Nigerian Association of Mathematical Physics, vol. 42, pp. 219-228 (2017).
[2] S. Shafiei and M. Doostparast, “Balakrishnan skew-t distribution and associated statistical characteristics,” Comm. Statist. Theory Methods, vol. 43, pp. 4109-4122 (2014).
[3] K. Aas and I. H. Haff, “The generalised hyperbolic skew Student’s t-distribution,” Journal of Financial Econometrics, vol. 4, no. 2, pp. 275-309 (2006).
[4] K. S. Khamis, D. Basalamah, W. Ning, and A. Gupta, “The Kumaraswamy skew-t distribution and its related properties,” Communications in Statistics - Simulation and Computation, vol. 47, no. 8, pp. 2409-2423 (2017), doi: 10.1080/03610918.2017.1346801.
[5] O. I. Shittu, K. A. Adepoju, and O. E. Adeniji, “On the beta skew-t distribution in modelling stock return in Nigeria,” International Journal of Modern Mathematical Sciences, vol. 11, no. 2, pp. 94-102 (2014).
[6] D. Basalamah, W. Ning, and A. Gupta, “The beta skew-t distribution and its properties,” Journal of Statistical Theory and Practice, vol. 12, pp. 837-860 (2018).
[7] O. D. Adubisi, A. Abdulkadir, and H. Chiroma, “Type-I half logistic skew-t distribution: A heavy-tail model with inverted bathtub shaped hazard rate,” Scientific African, submitted for publication (2020).
[8] M. C. Jones and M. J. Faddy, “A skew extension of the t distribution, with applications,” J. Roy. Statist. Soc., Ser. B, vol. 65, no. 2, pp. 159–174 (2003).
[9] G. M. Cordeiro, M. Alizadeh, and P. R. D. Marinho, “The Type-I half-logistic family of distributions,” Journal of Statistical Computation and Simulation, vol. 86, no. 1, pp. 707–728 (2015).
[10] F. Louzada, P. L. Ramos, and G. S. C. Perdoná, “Different estimation procedures for the parameters of the extended exponential geometric distribution for medical data,” Computational and Mathematical Methods in Medicine, vol. 2016, Article ID 8727951, pp. 1-12 (2016), doi: 10.1155/2016/8727951.
[11] G. C. Rodrigues, F. Louzada, and P. L. Ramos, “Poisson–exponential distribution: different methods of estimation,” Journal of Applied Statistics, vol. 45, no. 1, pp. 128–144 (2018).
[12] H. S. Bakouch, S. Dey, P. L. Ramos, and F. Louzada, “Binomial-exponential 2 distribution: Different estimation methods with weather applications,” TEMA (São Carlos), vol. 18, no. 2, pp. 233–251 (2017).
[13] C. Chesneau, H. S. Bakouch, P. L. Ramos, and F. Louzada, “The polynomial-exponential distribution: A continuous probability model allowing for occurrence of zero values,” Communications in Statistics-Simulation and Computation, vol. 99, pp. 1–26 (2020), doi: 10.1080/03610918.2020.1746339.
[14] P. L. Ramos, F. Louzada, E. Ramos, and S. Dey, “The Fréchet distribution: Estimation and application-An overview,” Journal of Statistics and Management Systems, vol. 23, no. 3, pp. 549–578 (2020).
[15] M. A. D. Aldahlan and A. M. Afify, “The odd exponential half-logistic exponential distribution: Estimation methods and application to Engineering data,” Mathematics, vol. 8, no. 10, pp. 1–33 (2020), doi: 10.3390/math8101684.
[16] R. A. ZeinEldin, C. Chesneau, F. Jamal, and M. Elgarhy, “Different estimation methods for Type-I half-logistic Topp-Leone distribution,” Mathematics, vol. 7, no. 10, pp. 1–23 (2019), doi: 10.3390/math7100985.
[17] R. Dewi, H. Wu, Z. Z., and A. D. T., “Maximum likelihood estimation for the proportion difference of two-sample binomial data subject to one type of misclassification,” Journal of Statistics and Management Systems, vol. 22, no. 8, pp. 1365–1379 (2019), doi: 10.1080/09720510.2019.1606319.
[18] Rui Cheng and Nasir Amin, “Maximum product of spacing estimation with application to Lognormal distribution,” Mathematical Report 79-1, University of Wales, Cardiff, UK (1979).
[19] Rui Cheng and Nasir Amin, “Estimating parameters in continuous univariate distributions with a shifted origin,” J. R. Stat. Soc. Ser. B Methodol., vol. 45, pp. 394-403 (1983).
[20] Bertil Ranneby, “The maximum spacing method: An estimation method related to the maximum likelihood method,” Scand. J. Stat., vol. 11, pp. 93-112 (1984).
[21] Alberto Lucceno, “Fitting the generalized Pareto distribution to data using maximum goodness-of-fit estimators,” Comput. Stat. Data Anal., vol. 51, pp. 904-917 (2006).
[22] Peter MacDonald, “Comment on an estimation procedure for mixtures of distribution by Choi and Bulgren,” J. R. Stat. Soc. Ser. B Methodol., vol. 29, pp. 271-329 (1971).
[23] James J. Swain, S. Venkatraman, and James R. Wilson, “Least-squares estimation of distribution functions in Johnson’s translation system,” J. Stat. Comput. Simul., vol. 29, pp. 271–297 (1988).

Views: 94Downloads: 64Citations: 0