Comparison some censored regression models with application to renal failure data

Document Type : Research Paper

Authors

1 College of Administration and Economics, University of Kirkuk, Iraq

2 College Education for Pure Sciences, University of Kirkuk, Iraq

Abstract

This article dealt with the topic of using some functions in the censored regression model, such as the Tobit model and Log-BurrXIIEE model and estimating the parameters of the two models. Based on real data for patients with renal failure, the two models were compared. Results are showed  that the LBXIIEE model is the best model compared to the Tobit model according to the values of the model selection criteria, as the values of these criteria for the LBXIIEE model were less than the values of the same criteria for the Tobit model when Urea as a dependent variable in the model. We also note that the values of the criteria (AIC, BIC, H-QIC) change and decrease according to the steps used for estimation and analysis of the two models (Tobit, LBXIIEE) according to the specific dependent variable

Keywords

[1] Al-Rubaie, M. A. (2019). “Some nonparametric capabilities of the first type of observational data.” Master’s
Thesis - College of Administration and Economics - University of Baghdad.
[2] Bang, H., & Tsiatis, A. A. (2000), Estimating medical costs with censored data, Biometrika,87(2), 329-343.
[3] Carson, R. T., & Sun, Y. (2007). The Tobit model with a non-zero threshold. The Econometrics Journal, 10(3),
488-502.
[4] Chen, E. H., & Dixon, W. J. (1972). Estimates of parameters of a censored regression sample. Journal of the
American Statistical Association, 67(399), 664-671.
[5] Chib, S. (1992). Bayes inference in the Tobit censored regression model. Journal of Econometrics, 51(1-2), 79-99.
[6] Cordeiro, G. M., Yousof, H. M., Ramires, T. G., & Orte ga, E. M. (2018). The Burr XII system of densities:
properties, regression model and applications. Journal of Statistical Computation and Simulation, 88(3), 432-456.
[7] Gupta, R. D., & Kundu, D. (2001). Exponentiated exponential family: an alternative to gamma and Weibull
distributions. Biometrical Journal: Journal of Mathematical Methods in Biosciences, 43(1), 117-130.
[8] Hadi, F. H. 2017. “Studying the variables affecting the number of active sperms using the Tobit Model.” AlQadisiyah University - College of Administration and Economics - Statistics Department.
[9] Ibrahim, M., Ea, E. A., & Yousof, H. M. (2020). A new distribution for modeling lifetime data with different
methods of estimation and censored regression modeling. Statistics, Optimization & Information Computing, 8(2),
610-630.
[10] Karim, A. A. (2018) “Comparison between the contraction method and the maximum likelihood method for
estimating the survival function of the Weibull distribution in the case of type I censored data using simulation.”
Master’s Thesis in Statistics - College of Administration and Economics - University of Karbala.[11] Parana´ıba, P. F., Ortega, E. M., Cordeiro, G. M., & Pescim, R. R. (2011). The beta Burr XII distribution with
application to lifetime data. Computational Statistics & Data Analysis, 55(2), 1118-1136.
[12] Z., Lin, D. Y., & Ying, Z. (2006). On least-squares regression with censored data. Biometrika, 93(1), 147-161.
Volume 12, Special Issue
December 2021
Pages 799-807
  • Receive Date: 17 February 2021
  • Revise Date: 05 July 2021
  • Accept Date: 20 August 2021