Survival and Reliability Analysis with an Epsilon-Positive Family of Distributions with Applications

We introduce a new class of distributions called the epsilon–positive family, which can be viewed as generalization with positive support. The construction family is motivated by ideas behind generation skew using symmetric kernels. This has special cases exponential, Weibull, log–normal, log–logistic and gamma distributions, it provides an alternative for analyzing reliability survival data. An interesting feature that finite scale mixture facilitating derivation implementation EM–type algorithms to obtain maximum likelihood estimates (MLE) (un)censored illustrate flexibility this analyze censored uncensored data two real examples. One them was previously discussed in literature; second one consists application model recidivism group inmates released from Chilean prisons during 2007. results show better performance fitting than some common alternatives such exponential distribution.