Residual Lifetime Distributions and their Applications: The Median
Master’s Candidate, Department of Statistics, Colorado State University
Tuesday, 14 October 2003
Suppose that we want to buy a used item such as computer. In such a
situation we do not know the age (hours in working) of the computer. Therefore, it would be of interest
to have some criteria, which can be used to give us indications of the
future performance of the purchased item. The concept of aging has been
discussed in terms of the individual's survival time distribution such as
hazard rate, hazard average, and mean residual lifetime. However, in some
instances the hazard rate or the average hazard rate does not give a clear
view, and in some instances the mean residual lifetime does not even exist.
In this study we consider the median residual lifetime for some
distributions as describing aging. Explicit asymptotic expression for the
median of corresponding residual lifetime distributions are found. Finally
these expressions are empirically calculated and plotted.