Prediction Intervals for Generalized Burr Regression Models

Document Type : Original Article

Author

Abstract

In this paper, equal-tailed two-sided prediction intervals are developed for generalized Burr type XII, or log-Burr regression model. A useful parameterization given by Gould and Lawless (1988) is used to write the log-Burr model as a location-scale parameter regression model. An iterative procedure is needed to obtain both limits. The case of random sample from a single generalized Burr type XII, or log-Burr distribution is considered as a special case. In this case the results can be extended to include type II censored sampling.

Keywords