Abstract
We propose the exponentiated odd log-logistic normal quantile regression model relating the covariates to the parameters through two systematic components, and adopt the maximum likelihood method to estimate the parameters. We report Monte Carlo simulations to verify the accuracy of the maximum likelihood estimators. The usefulness of the new regression is proved using three applications to real data. It and its special cases proved to be interesting alternatives to model data in different shapes, and obtain important information on the response variable from the analysis of different quantiles.
Acknowledgments
The authors are very grateful to the editor and two referees for helpful comments.
Disclosure statement
No potential conflict of interest was reported by the author(s).