Advances and Applications in Statistics
Volume 53, Issue 2, Pages 87 - 102
(August 2018) http://dx.doi.org/10.17654/AS053020087 |
|
GENERALIZED WEIBULL LINEAR MODELS FOR TYPE-I CENSORING DATA
Afsana-Al-Sharmin and M. Ataharul Islam
|
Abstract: For the first time, a new class of generalized Weibull linear models is introduced for the analysis of Type-I censored data. In this article, the parameter estimation method of Weibull regression model is discussed by utilizing techniques of generalized linear model (GLM). Four generalized Weibull linear models (GWLM) are proposed in this study. The likelihood functions for the models of Type-I censoring are shown for the Weibull distribution considering four different link functions. The iteratively reweighted least squares method is implemented to obtain the maximum likelihood estimation solution. This development enhances the ability of Weibull distribution to solve the problem of analyzing Type-I censoring data on the ground of GLM. We have done some simulations for making a comparison among four proposed models and found out the superior one. |
Keywords and phrases: generalized Weibull linear model, negative link, log link, identity link, reparameterization, Type-I censoring. |
|
Number of Downloads: 441 | Number of Views: 3446 |
|