Heteroscedastic log-exponentiated Weibull regression model

ABSTRACT

We introduce a new class of heteroscedastic log-exponentiated Weibull (LEW) regression models. The class of regression models can be applied to censored data and be used more effectively in survival analysis. Maximum likelihood estimation of the model parameters with censored data as well as influence diagnostics for the new regression model is investigated. For different parameter settings, sample sizes and censoring percentages, various simulation studies are performed and compared to the performance of the heteroscedastic LEW regression model. The normal curvatures for studying local influence are derived under various perturbation schemes. An empirical application to a real data set is provided to illustrate the usefulness of the new class of heteroscedastic regression models.

KEYWORDS:

• Exponentiated Weibull distribution
• lifetime data
• location-scale regression model
• log-Weibull regression model

MATHEMATICS SUBJECT CLASSIFICATION:

• 62F10
• 62J05
• 62J20

Acknowledgements

The authors would like to express thanks to two anonymous referees for useful suggestions and comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (Brazil).