A unit is said to be interval-censored if it is known that the event of interest lies between two values (time), but the exact time of failure is unknown. Both frequentist-based and Bayesian estimation procedure are considered when the data under consideration is interval-censored using a generalized non-informative prior, which contains Jeffreys prior as a special case. The study is based on simulation and the comparisons are made using mean squared errors and absolute bias. A set of real data is also analyzed to illustrate the methods proposed in this paper.

interval-censoring, maximum likelihood, Bayesian estimation, generalized non-informative prior.