摘要

In recent advances of Weibull generalization, the odd Weibull family has been shown to be useful not only for lifetime data modeling, but also discriminating between Weibull and inverse Weibull distributions. This three-parameter distribution accommodates seven different hazard shapes and a wide variety of shapes of the density function including bimodality. In addition, the odd Weibull parameters can be estimated in two different ways since the inverse transformation of the family does not change its density function. In this article, we adapted this two-way estimation method for analyzing grouped, censored, and truncated data that frequently encountered in survival analysis.

  • 出版日期2012